{
  "title": "2026-06-20 AI Daily | Agents Enter the Governance Era, Edge Models Begin to Reshape Workflows",
  "url": "https://miaok.ong/en/ai-daily/ai-daily-2026-06-20/",
  "date": "2026-06-20T07:00:00+08:00",
  "lastmod": "2026-06-20T07:00:00+08:00",
  "type": "ai-daily",
  "kind": "page",
  "language": "en",
  "description": "Today\u0026rsquo;s main theme shifts from model capabilities to controllable implementation: agents require permissions, auditing, and runtime governance; edge models are rapidly maturing in multimodal, programming, and personal assistant scenarios; and AI coding tools are evolving from assistive generation to reusable automated processes.",
  "keywords": null,
  "tags": [],
  "categories": [],
  "author": "Mark (Miao) Kong",
  "image": "https://miaok.ong/images/avatar.jpg",
  "content": "\u003ch1 id=\"2026-06-20-ai-daily--agents-enter-the-governance-era-on-device-models-start-reshaping-workflows\"\u003e\n  2026-06-20 AI Daily | Agents Enter the Governance Era, On-Device Models Start Reshaping Workflows\n  \u003ca class=\"heading-link\" href=\"#2026-06-20-ai-daily--agents-enter-the-governance-era-on-device-models-start-reshaping-workflows\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h1\u003e\n\u003cblockquote\u003e\n\u003cp\u003eToday\u0026rsquo;s main theme shifts from model capabilities to controllable implementation: Agents require permissions, auditing, and runtime governance; on-device models are rapidly maturing in multimodal, programming, and personal assistant scenarios; and AI coding tools are evolving from assisted generation towards reusable automated workflows.\u003c/p\u003e\n\u003c/blockquote\u003e\n\u003ch2 id=\"-in-depth-guide-to-this-issues-watch-list\"\u003e\n  📖 In-depth Guide to This Issue\u0026rsquo;s Watch List\n  \u003ca class=\"heading-link\" href=\"#-in-depth-guide-to-this-issues-watch-list\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003cp\u003eThe most important theme to explore today is \u0026ldquo;Agents Moving from Demonstration to Governance.\u0026rdquo; The interview with the founder of OpenClaw is highly relevant for product and engineering teams: truly valuable Agents are being defined by domain experts who understand the context. Meanwhile, several papers on runtime governance, DeFi risk supervision, and clarifying uncertainty remind us that systems capable of tool use must be integrated into a framework of permissions, obligations, and auditing.\u003c/p\u003e\n\u003cp\u003eThe second theme is model paradigms and reliability. Work like Diffusion Language Models and ITNet continues to challenge the default assumptions of Transformers/auto-regression. \u0026ldquo;Hidden anchors\u0026rdquo; in multi-agent deliberation, cognitive blind spots in clinical tabular data, and bias revealed through random path aggregation all point to one problem: models not only need to be more powerful, but they also need to know why they are wrong and when to stop.\u003c/p\u003e\n\u003cp\u003eFinally, the power structures within the tech industry also warrant attention. From the SpaceX IPO and rumors of Cursor\u0026rsquo;s acquisition to the controversy over Anthropic\u0026rsquo;s fables, the AI industry is simultaneously reshaping capital, platforms, and the public narrative.\u003c/p\u003e\n\u003ch2 id=\"-ai-hot-topics-on-x\"\u003e\n  🌐 AI Hot Topics on X\n  \u003ca class=\"heading-link\" href=\"#-ai-hot-topics-on-x\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003ch3 id=\"topic-1-transformer-pioneer-noam-shazeer-leaves-google-for-openai\"\u003e\n  Topic 1: Transformer Pioneer Noam Shazeer Leaves Google for OpenAI\n  \u003ca class=\"heading-link\" href=\"#topic-1-transformer-pioneer-noam-shazeer-leaves-google-for-openai\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eCategory: AI · News\u003c/li\u003e\n\u003cli\u003eOverview: Trending since: 1 day ago, Related posts: 15,000\u003c/li\u003e\n\u003cli\u003eWhat happened: Noam Shazeer, co-author of the Transformer paper and a pioneer of MoE, has reportedly left Google to join OpenAI, sparking widespread attention on the movement of top AI talent.\u003c/li\u003e\n\u003cli\u003eWhy it matters: Shazeer was instrumental in laying the groundwork for modern large model architectures. His move is considered a significant signal for measuring cutting-edge R\u0026amp;D capabilities, organizational attractiveness, and the future direction of AI technology.\u003c/li\u003e\n\u003cli\u003eDiscussion summary: Discussions on X are focused on whether Google is continuing to lose key AI talent, whether OpenAI is further solidifying its research lead, and who will pioneer the next-generation architecture or training paradigm that will succeed the Transformer. Some also questioned the accuracy of the report\u0026rsquo;s details and timeline.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"topic-2-loop-engineering-turns-ai-agents-into-self-sustaining-coders\"\u003e\n  Topic 2: Loop Engineering Turns AI Agents into Self-Sustaining Coders\n  \u003ca class=\"heading-link\" href=\"#topic-2-loop-engineering-turns-ai-agents-into-self-sustaining-coders\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eCategory: AI · News\u003c/li\u003e\n\u003cli\u003eOverview: Trending since: 10 hours ago, Related posts: 336\u003c/li\u003e\n\u003cli\u003eWhat happened: Loop Engineering has proposed a method that allows AI agents to autonomously write, test, and improve code through a cyclical development process, referring to them as \u0026ldquo;self-sustaining coders.\u0026rdquo;\u003c/li\u003e\n\u003cli\u003eWhy it matters: This reflects a trend of AI programming agents evolving from assisting with code generation to continuously performing engineering tasks. This could affect software development efficiency, the level of automation, and the division of roles for human engineers.\u003c/li\u003e\n\u003cli\u003eDiscussion summary: Discussions on X are mainly focused on whether such agents can truly achieve long-term autonomous development, how code quality and security can be guaranteed, and whether they will boost developer productivity or introduce risks of excessive automation and job displacement.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"topic-3-anthropic-fixes-claude-code-usage-bug-for-premium-users\"\u003e\n  Topic 3: Anthropic Fixes Claude Code Usage Bug for Premium Users\n  \u003ca class=\"heading-link\" href=\"#topic-3-anthropic-fixes-claude-code-usage-bug-for-premium-users\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eCategory: AI · News\u003c/li\u003e\n\u003cli\u003eOverview: Trending since: 20 hours ago, Related posts: 3,700\u003c/li\u003e\n\u003cli\u003eWhat happened: Anthropic has fixed a bug that affected usage statistics or credit deductions for premium subscribers of Claude Code.\u003c/li\u003e\n\u003cli\u003eWhy it matters: Claude Code is designed for high-frequency developer workflows, and the accuracy of its credit and billing systems is crucial for user trust, enterprise adoption, and the viability of its business model.\u003c/li\u003e\n\u003cli\u003eDiscussion summary: Discussions on X centered on whether Anthropic will promptly compensate affected users, whether Claude Code\u0026rsquo;s usage policies are sufficiently transparent, and if Anthropic can maintain developer confidence as competitors like OpenAI accelerate the launch of their programming agent capabilities.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"topic-4-zais-glm-52-tops-open-ai-model-charts-with-strong-benchmarks\"\u003e\n  Topic 4: Z.ai\u0026rsquo;s GLM-5.2 Tops Open AI Model Charts with Strong Benchmarks\n  \u003ca class=\"heading-link\" href=\"#topic-4-zais-glm-52-tops-open-ai-model-charts-with-strong-benchmarks\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eCategory: AI · News\u003c/li\u003e\n\u003cli\u003eOverview: Trending since: 1 day ago, Related posts: 19,000\u003c/li\u003e\n\u003cli\u003eWhat happened: GLM-5.2, released by Zhipu\u0026rsquo;s Z.ai, has achieved leading results on several open model benchmarks, attracting attention from the AI community.\u003c/li\u003e\n\u003cli\u003eWhy it matters: This demonstrates that China\u0026rsquo;s open-source large models are continuing to approach or surpass mainstream international models in reasoning, coding, and general capabilities, which could accelerate competition and application deployment in the open model ecosystem.\u003c/li\u003e\n\u003cli\u003eDiscussion summary: Discussions on X are mainly focused on whether the benchmark results reflect actual capabilities, the gap between GLM-5.2 and models like DeepSeek, Qwen, and Llama, and its attractiveness to developers regarding open weights, commercial licensing, and deployment costs.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch4 id=\"summary-of-todays-ai-sentiments-on-x\"\u003e\n  Summary of Today\u0026rsquo;s AI Sentiments on X\n  \u003ca class=\"heading-link\" href=\"#summary-of-todays-ai-sentiments-on-x\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h4\u003e\n\u003cp\u003eToday\u0026rsquo;s main narrative revolves around the concentration of cutting-edge AI capabilities in stronger organizations, more autonomous tools, and more open ecosystems. The movement of top research talent is seen as a bellwether for the competitiveness of institutions like OpenAI and Google, while advancements in coding agents and open-source models indicate that AI is shifting from a competition of model capabilities to one of practical engineering productivity. The consensus is that AI programming, reasoning, and the open-source model ecosystem are all rapidly maturing. Developer workflows will be profoundly reshaped, and platform advantages will be determined by a combination of talent, computing power, product experience, and business trust. The main points of contention are whether related news and benchmarks are reliable, whether autonomous coding agents truly possess long-term engineering capabilities, and whether the leading performance of Chinese open-source models can be translated into a stable advantage in real-world scenarios. Potential risks are concentrated in over-reliance on insufficiently validated agent systems, a lack of transparency in billing and quotas that erodes user trust, and the further concentration of talent and technical resources in a few leading institutions, which amplifies industry competition and governance pressures.\u003c/p\u003e\n\u003ch2 id=\"-influencer-insights\"\u003e\n  💡 Influencer Insights\n  \u003ca class=\"heading-link\" href=\"#-influencer-insights\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003ch1 id=\"ai-daily-industry-dynamics-analysis\"\u003e\n  AI Daily Industry Dynamics Analysis\n  \u003ca class=\"heading-link\" href=\"#ai-daily-industry-dynamics-analysis\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h1\u003e\n\u003cp\u003e\u003cstrong\u003eDate: June 18-19, 2026 (Synthesizing recent trends)\u003c/strong\u003e\u003c/p\u003e\n\u003ch2 id=\"1-todays-focus-the-rise-of-on-device-models-and-their-engineering-practices\"\u003e\n  1. Today\u0026rsquo;s Focus: The Rise of On-device Models and Their Engineering Practices\n  \u003ca class=\"heading-link\" href=\"#1-todays-focus-the-rise-of-on-device-models-and-their-engineering-practices\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003cp\u003eOver the past 24 hours, discussions among AI leaders have strongly pointed to a core trend: \u003cstrong\u003eHigh-performance on-device models are transitioning from \u0026ldquo;usable\u0026rdquo; to \u0026ldquo;great to use\u0026rdquo; and are beginning to reshape the workflows of developers and power users.\u003c/strong\u003e\u003c/p\u003e\n\u003ch3 id=\"11-on-device-model-capabilities-validated-with-breakthroughs-in-both-performance-and-quality\"\u003e\n  1.1 On-device Model Capabilities Validated, with Breakthroughs in Both Performance and Quality\n  \u003ca class=\"heading-link\" href=\"#11-on-device-model-capabilities-validated-with-breakthroughs-in-both-performance-and-quality\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e@zhixianio\u003c/strong\u003e expressed being \u0026ldquo;very satisfied\u0026rdquo; with the full-duplex audio-visual performance of \u003cstrong\u003eMiniCPM-o 4.5\u003c/strong\u003e, marveling, \u0026ldquo;It\u0026rsquo;s hard to imagine a 9B model can achieve this.\u0026rdquo; Although stability issues persist during long runs, its quality is already initially viable.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e@zhixianio\u003c/strong\u003e dubbed \u003cstrong\u003eQwen3.6-35B-A3B MoE (oMLX)\u003c/strong\u003e the \u0026ldquo;sweet spot 🍮 champion,\u0026rdquo; noting its response speed in Personal Assistant (PA) and Coding scenarios surpasses remote LLMs, and its native multimodal user experience is \u0026ldquo;even better than DSV4 Pro.\u0026rdquo;\u003c/li\u003e\n\u003cli\u003eThe \u003cstrong\u003eGoogle Gemma 4 family\u003c/strong\u003e has become a focus for on-device applications:\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003eGemma 4 E4B + MTP\u003c/strong\u003e: @zhixianio\u0026rsquo;s tests confirmed its excellent performance on Japanese email parsing and classification tasks.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eGemma 4 12B Coder\u003c/strong\u003e: @zhixianio conducted rigorous comparative tests on code generation. The conclusion was that in a head-to-head with \u003cstrong\u003eQwen3.6-35B-A3B MoE\u003c/strong\u003e, Gemma 12B Coder hit a clear \u0026ldquo;ceiling\u0026rdquo; in generating complex, stateful programs (like Tetris), as its 12B parameters struggled to support long-form complex logic.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eGemma 4 QAT Quantization-Aware Model\u003c/strong\u003e: @zhixianio specifically highlighted Google\u0026rsquo;s Quantization-Aware Training (QAT) approach, considering it a key direction for on-device optimization, adding, \u0026ldquo;Android will soon be able to use its own native models.\u0026rdquo;\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"12-ai-coding-tools-enter-a-clash-of-titans-and-a-new-stage-of-automation\"\u003e\n  1.2 AI Coding Tools Enter a \u0026ldquo;Clash of Titans\u0026rdquo; and a New Stage of Automation\n  \u003ca class=\"heading-link\" href=\"#12-ai-coding-tools-enter-a-clash-of-titans-and-a-new-stage-of-automation\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003eClaude Code vs. OpenAI Codex\u003c/strong\u003e: @ruanyf\u0026rsquo;s question, \u0026ldquo;Do you use Codex or Claude Code?\u0026rdquo; sparked a discussion. @vista8 stated that \u003cstrong\u003eCodex\u003c/strong\u003e is the superior product, but specific scenarios still require \u003cstrong\u003eClaude Code\u003c/strong\u003e. They use a self-developed MCP to enable synergy between the two, even achieving a \u0026ldquo;double Codex quota\u0026rdquo; hack. @gefei55 also shared an open-source project that uses an MCP to give the ChatGPT web version local code manipulation capabilities via Codex, which is also essentially aimed at \u0026ldquo;doubling the quota\u0026rdquo; and calling the most powerful model.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eClaude Code launches Artifacts for visual collaboration\u003c/strong\u003e: @dotey provided a detailed interpretation of this feature, stating that it solves the collaboration problem where AI programming results are \u0026ldquo;only visible to the operator.\u0026rdquo; It allows work products from debugging, PR reviews, and architectural explanations to be shared with the team as real-time web pages.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eOpenAI Codex\u0026rsquo;s \u0026ldquo;Record \u0026amp; Replay\u0026rdquo; feature\u003c/strong\u003e: @dotey and @AI_Jasonyu highly praised this feature, viewing it as essentially a \u0026ldquo;super version of RPA + keyboard macro + Computer Use combined.\u0026rdquo; Users only need to demonstrate a workflow once, and Codex can automatically generate a reusable Skill, significantly lowering the barrier to automation.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"2-noteworthy-unique-perspectives-and-industry-foresight\"\u003e\n  2. Noteworthy Unique Perspectives and Industry Foresight\n  \u003ca class=\"heading-link\" href=\"#2-noteworthy-unique-perspectives-and-industry-foresight\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003ch3 id=\"21-the-technological-paradigm-shift-from-correlation-to-causality\"\u003e\n  2.1 The Technological Paradigm Shift from \u0026ldquo;Correlation\u0026rdquo; to \u0026ldquo;Causality\u0026rdquo;\n  \u003ca class=\"heading-link\" href=\"#21-the-technological-paradigm-shift-from-correlation-to-causality\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e@Pluvio9yte\u003c/strong\u003e offered a deep analysis of \u003cstrong\u003eAether AI\u003c/strong\u003e, founded by Professor Biwei Huang, and pointed out that its \u0026ldquo;Causal World Models\u0026rdquo; indicate a key direction for the next stage. He argues that current large models fail when generating images like \u0026ldquo;pouring water into a cup with a hole in the bottom\u0026rdquo; because they learn the data correlation between \u0026ldquo;pouring water\u0026rdquo; and a \u0026ldquo;full cup,\u0026rdquo; not the physical causal mechanism that \u0026ldquo;water leaks from a hole.\u0026rdquo; Aether AI aims to enable AI to understand environmental variables and intervention outcomes, which is crucial for fields demanding rigorous logic, such as embodied intelligence and new materials R\u0026amp;D.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"22-the-evolution-of-vibe-coding-from-requirements-first-to-contract-first\"\u003e\n  2.2 The Evolution of Vibe Coding: From \u0026ldquo;Requirements-First\u0026rdquo; to \u0026ldquo;Contract-First\u0026rdquo;\n  \u003ca class=\"heading-link\" href=\"#22-the-evolution-of-vibe-coding-from-requirements-first-to-contract-first\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e@Pluvio9yte\u003c/strong\u003e shared his in-depth thoughts on transitioning from a security professional to a full-stack developer. He argues that the best practice for \u0026ldquo;Vibe Coding\u0026rdquo; is not \u003cstrong\u003eRequirement First\u003c/strong\u003e or \u003cstrong\u003eCode First\u003c/strong\u003e, but \u003cstrong\u003eContract First\u003c/strong\u003e. By combining this experience with the open-source project OpenSpec, he has developed a framework that \u0026ldquo;externalizes easily shifting context into contracts, providing a stable reference for both humans and AI.\u0026rdquo; The goal is to enable less experienced developers to systematically undertake large-scale project development.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"23-personal-development-philosophy-in-the-ai-era\"\u003e\n  2.3 Personal Development Philosophy in the AI Era\n  \u003ca class=\"heading-link\" href=\"#23-personal-development-philosophy-in-the-ai-era\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e@ruanyf\u003c/strong\u003e shared the article \u0026ldquo;Can I Take a Day Off Today?\u0026rdquo;, questioning where the benefits for employees lie after AI significantly boosts the productivity of white-collar workers. He predicts that AI will increase the average salary or welfare of the entire society as a long-term trend.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e@lijigang\u003c/strong\u003e, drawing from the philosophy that \u0026ldquo;Bodhisattvas fear the cause, while ordinary people fear the effect,\u0026rdquo; points out that an individual\u0026rsquo;s \u003cstrong\u003ethree core views (view of life, world, and values) are the fundamental \u0026lsquo;function f\u0026rsquo;\u003c/strong\u003e for handling life\u0026rsquo;s events. Designing this function well is superior to praying for a single outcome (f(x)). He also reminds us that \u003cstrong\u003e\u0026ldquo;Token consumption is a \u0026lsquo;vanity metric,\u0026rsquo; while the effectiveness in solving problems is the \u0026rsquo;true metric.\u0026rsquo;\u0026rdquo;\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e@gefei55\u003c/strong\u003e shared an AI-assisted business intelligence strategy for gaining a time advantage: using the X API to filter recent, high-engagement tweets with links to discover new terms and products before they signal on Google Trends. This achieves a state where \u0026ldquo;while others are waiting for the Trends curve, you\u0026rsquo;re already ahead of the game.\u0026rdquo; He has also open-sourced this method.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"3-recommended-tools--resources\"\u003e\n  3. Recommended Tools \u0026amp; Resources\n  \u003ca class=\"heading-link\" href=\"#3-recommended-tools--resources\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003ch3 id=\"31-development--design-tools\"\u003e\n  3.1 Development \u0026amp; Design Tools\n  \u003ca class=\"heading-link\" href=\"#31-development--design-tools\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003ebaoyu-design Skill (by @dotey)\u003c/strong\u003e: A powerful local design Skill that supports generating animated videos and exporting them as MP4s. It can also generate a PPTX with images in one click, which can be further edited. Its animation engine uses a declarative design \u003ccode\u003ef(t)\u003c/code\u003e and supports frame-accurate exporting. The GitHub address has been released.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eMeta Skill 2.0 (by @yaojingang, recommended by @vista8)\u003c/strong\u003e: A \u0026ldquo;meta-Skill\u0026rdquo; created by integrating the leaked Claude Code source code from Anthropic. It\u0026rsquo;s used for creating high-quality Skills and was called \u0026ldquo;the best Meta Skill I\u0026rsquo;ve ever used\u0026rdquo; by @vista8.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eFigma Chrome Plugin (recommended by @vista8)\u003c/strong\u003e: Can convert any web page element into an editable layer in Figma, which is extremely useful for designers replicating websites and taking precise screenshots.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eFable (recommended by @zhixianio)\u003c/strong\u003e: An AI-driven development assistant that can complete 70% of a demo in 40 minutes and optimize the original plan. It was reviewed as \u0026ldquo;Shut up and take my money.\u0026rdquo;\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"32-productivity--automation\"\u003e\n  3.2 Productivity \u0026amp; Automation\n  \u003ca class=\"heading-link\" href=\"#32-productivity--automation\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003eYouMind 1.0 (by @lifesinger, recommended by @AI_Jasonyu, @gefei55)\u003c/strong\u003e: A content creation tool officially released after two years of development. Its highlights are high-quality article output and excellent formatting compatibility with platforms like X and WeChat Official Accounts. Combined with its image-adding feature, it offers a one-stop solution for the pain points of long-form content creation.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eLocal Video Translation Tool (by @xiaohu, recommended by @Pluvio9yte)\u003c/strong\u003e: An open-source, all-in-one video translation tool that automates the entire process: downloading, transcription, translation, polishing, and burning subtitles.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eEvoMap (recommended by @vista8)\u003c/strong\u003e: A platform event where Stars on GitHub open-source projects can be exchanged for large model API Tokens. It encourages developers to package workflows and Prompts and upload them to earn more rewards.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eGiffgaff UK SIM Card Guide (by @AI_Jasonyu)\u003c/strong\u003e: A detailed tutorial for solving the phone number verification challenges for overseas AI services (like Codex, Claude Code).\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"33-cutting-edge-models--frameworks\"\u003e\n  3.3 Cutting-Edge Models \u0026amp; Frameworks\n  \u003ca class=\"heading-link\" href=\"#33-cutting-edge-models--frameworks\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003eoMLX v0.4.0 (released by @jundotkim, shared by @zhixianio)\u003c/strong\u003e: The first official version to support native Swift macOS apps, making it a key tool for smoothly running MLX models.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003ePP-OCRv6 (by @AI_Jasonyu)\u003c/strong\u003e: An ultra-lightweight OCR model released by Baidu. At only 1.5MB, it can run in a browser, process a single image in as fast as 97ms, and its character-by-character recognition accuracy surpasses large models like GPT-5.5, making it highly suitable for on-device integration.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eOpenClaw (mentioned by @zhixianio)\u003c/strong\u003e: A personal assistant framework that integrates local models. Paired with on-device models (like Qwen), it enables fast and intelligent native multi-modal interactions.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"-appendix-todays-watch-list-source-updates\"\u003e\n  📚 Appendix: Today\u0026rsquo;s Watch List Source Updates\n  \u003ca class=\"heading-link\" href=\"#-appendix-todays-watch-list-source-updates\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h2\u003e\n\u003cblockquote\u003e\n\u003cp\u003eTime window: Last 3 days; 22 sources covered; 34 updates in total\u003c/p\u003e\n\u003c/blockquote\u003e\n\u003ch3 id=\"y-combinator-podcast-b_introsearch\"\u003e\n  Y Combinator Podcast (B_intro+search)\n  \u003ca class=\"heading-link\" href=\"#y-combinator-podcast-b_introsearch\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e\u003ca href=\"https://podcasters.spotify.com/pod/show/ycombinator/episodes/Why-Domain-Experts-Are-Winning-In-The-Age-Of-AI-e3l08ia\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eWhy Domain Experts Are Winning In The Age Of AI\u003c/a\u003e\u003c/strong\u003e\n\u003cul\u003e\n\u003cli\u003eRelease Time: 2026-06-19 23:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - You may have heard of OpenClaw (formerly known as Clawdbot/Moltbot).\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eThe breakout open-source AI assistant that runs on your own device, connects with the messaging apps you already use, and goes beyond chat to actually perform tasks like managing email, calendars, files, workflows, and more.\u003c/li\u003e\n\u003cli\u003eNow, meet the person behind it.\u003c/li\u003e\n\u003cli\u003eYC\u0026rsquo;s Raphael Schaad sits down with OpenClaw founder Peter Steinberger to discuss the \u0026ldquo;aha\u0026rdquo; moment behind the viral personal AI agent, why a local-first agent could replace many of today\u0026rsquo;s apps, and how personal agents will reshape the future of software.\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003eBryant Chou co-founded Webflow, which today powers around 1% of all websites on the internet\u003c/li\u003e\n\u003cli\u003eNow he\u0026rsquo;s back in the current YC batch with Ploy, an AI-powered website and marketing platform that doesn\u0026rsquo;t just build your site — it connects to your analytics,…\u003c/li\u003e\n\u003cli\u003eIn this episode of the Lightcone he explains how he built Ploy to be “anti slop,” how building today compares to his first startup, and why founders with domain…\u003c/li\u003e\n\u003cli\u003eNow26:01 — First Three Months: Webflow 2013 vs\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"all-in-podcast-a_full\"\u003e\n  All-In Podcast (A_full)\n  \u003ca class=\"heading-link\" href=\"#all-in-podcast-a_full\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e\u003ca href=\"https://allinchamathjason.libsyn.com/worlds-first-trillionaire-anthropic-fable-banned-the-new-oligarchs-iran-peace-deal\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eWorld\u0026rsquo;s First Trillionaire, Anthropic Fable Banned, The New Oligarchs, Iran Peace Deal\u003c/a\u003e\u003c/strong\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-20 06:07 Beijing Time\u003c/li\u003e\n\u003cli\u003eSummary: - (0:00) Bestie intros.\n\u003cul\u003e\n\u003cli\u003e(2:41) The New Oligarchs, America\u0026rsquo;s incoming politburo, and learned helplessness.\u003c/li\u003e\n\u003cli\u003e(14:18) SpaceX\u0026rsquo;s record-breaking IPO, $60B Cursor acquisition, and trillionaire reactions.\u003c/li\u003e\n\u003cli\u003e(33:34) Behind the scenes of Anthropic\u0026rsquo;s Fable ban.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003e(0:00) Bestie intros\u003c/li\u003e\n\u003cli\u003e(2:41) The New Oligarchs, America\u0026rsquo;s incoming politburo, and learned helplessness\u003c/li\u003e\n\u003cli\u003e(14:18) SpaceX\u0026rsquo;s record breaking IPO, $60B Cursor acquisition, and trillionaire reactions\u003c/li\u003e\n\u003cli\u003e(33:34) Behind the scenes of Anthropic\u0026rsquo;s Fable ban\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"stratechery-by-ben-thompson-a_full\"\u003e\n  Stratechery by Ben Thompson (A_full)\n  \u003ca class=\"heading-link\" href=\"#stratechery-by-ben-thompson-a_full\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e\u003ca href=\"https://stratechery.com/2026/the-stuff-of-mythos/\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003e2026.25: The Stuff of Myth(os)\u003c/a\u003e\u003c/strong\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-20 01:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eSummary: - (Photo by Ronald Cortes/Getty Images).\n\u003cul\u003e\n\u003cli\u003eWelcome back to This Week in Stratechery!\u003c/li\u003e\n\u003cli\u003eAs a reminder, each week, every Friday, we’re sending out this overview of content in the Stratechery bundle; highlighted links are free for everyone.\u003c/li\u003e\n\u003cli\u003eAdditionally, you have complete control over what we send to you.\u003c/li\u003e\n\u003cli\u003eWith that said, here are some of our favorites from the week.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003e(Photo by Ronald Cortes/Getty Images)\u003c/li\u003e\n\u003cli\u003eWelcome back to This Week in Stratechery\u003c/li\u003e\n\u003cli\u003eAs a reminder, each week, every Friday, we’re sending out this overview of content in the Stratechery bundle; highlighted links are free for everyone\u003c/li\u003e\n\u003cli\u003eAdditionally, you have complete control over what we send to you\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"two-minute-papers-b_introsearch\"\u003e\n  Two Minute Papers (B_intro+search)\n  \u003ca class=\"heading-link\" href=\"#two-minute-papers-b_introsearch\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003e\u003ca href=\"https://www.youtube.com/watch?v=dUmT0OIGoqE\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eScientists Found A Better Language For AI Agents\u003c/a\u003e\u003c/strong\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 22:06 Beijing Time\u003c/li\u003e\n\u003cli\u003eSummary: - ❤️ Check out Weights \u0026amp; Biases and sign up for a free demo here:.\n\u003cul\u003e\n\u003cli\u003e📝 The paper is available here:.\u003c/li\u003e\n\u003cli\u003eAdam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan Benet, Michael Tedder, Owen Skarpness, Richard Sundvall, Ryan Stankye, Shawn Becker, Steef, Taras Bobrovytsky, Tazaur Sagenclaw, Tybie Fitzhugh, Ueli Gallizzi.\u003c/li\u003e\n\u003cli\u003eScientists have found a better language for AI agents.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003e❤️ Check out Weights \u0026amp; Biases and sign up for a free demo here:\u003c/li\u003e\n\u003cli\u003e📝 The paper is available here:\u003c/li\u003e\n\u003cli\u003eBrain reading video:\u003c/li\u003e\n\u003cli\u003e🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"arxiv-csai-b_introsearch\"\u003e\n  ArXiv cs.AI (B_intro+search)\n  \u003ca class=\"heading-link\" href=\"#arxiv-csai-b_introsearch\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19464\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eDeontic Policies for Runtime Governance of Agentic AI Systems\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eSummary: - arXiv:2606.19464v1 Announcement Type: New.\n\u003cul\u003e\n\u003cli\u003eAbstract: Autonomous agentic AI systems driven by Large Language Models (LLMs) introduce a new class of security, privacy, and compliance challenges: agents that can invoke tools, operate on data, install software, and coordinate with peer agents across organizational boundaries must be subject not only to authentication and access controls but also to the full structure of enterprise governance.\u003c/li\u003e\n\u003cli\u003eThis includes specifying what agents are permitted and prohibited from doing, what they are obliged to do after taking certain actions (e.g., notify the CISO), under what conditions long-term obligations can be waived, and which rules take precedence when policies conflict.\u003c/li\u003e\n\u003cli\u003eThis governance problem exceeds what current policy engines can provide.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19464v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Autonomous agentic AI systems driven by Large Language Models (LLMs) introduce a new class of security, privacy, and compliance challenges: an agent t…\u003c/li\u003e\n\u003cli\u003eThis includes specifying what agents are permitted and prohibited from doing, what they areobliged to do after certain actions (e.g., notify the CISO), under wh…\u003c/li\u003e\n\u003cli\u003eThis governance problem exceeds what current policy engines provide\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19469\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eMeasuring Curriculum Alignment across Topical Coverage, Competency, and Cognitive Depth: A Longitudinal Framework Applied to CS2013 and CS2023\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eSummary: - arXiv:2606.19469v1 Announcement Type: New.\n\u003cul\u003e\n\u003cli\u003eAbstract: Undergraduate computer science is governed by international curriculum guidelines that are revised approximately every ten years, yet curricula lack reliable, repeatable methods to measure the extent to which they cover the current guidelines and how that coverage changes when the guidelines are restructured.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eWe address this with a human-in-the-loop pipeline that measures a program\u0026rsquo;s coverage of an external body of knowledge, applied longitudinally to accredited computer science bachelor\u0026rsquo;s degrees according to the 2013 (CS2013) and 2023 (CS2023) computer science curricula.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eThe pipeline represents the program and each guideline as structured corpora, generates candidate course-to-knowledge-unit matches via semantic retrieval, and confirms them through human judgment under clear coverage definitions.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19469v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Undergraduate computer science is governed by international curricular guidelines revised about once a decade, yet programs lack a reliable, reproduci…\u003c/li\u003e\n\u003cli\u003eWe address this with a human-in-the-loop pipeline that measures a program\u0026rsquo;s coverage of an external body of knowledge, applied longitudinally to one accredited…\u003c/li\u003e\n\u003cli\u003eThe pipeline represents the program and each guideline as structured corpora, generates candidate course-to-knowledge-unit matches by semantic retrieval, and co…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19475\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eDiffusion Language Models: An Experimental Analysis\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19475v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Large Language Models (LLMs) have revolutionized language modeling through autoregressive generation, enabling strong performance across a wide range of tasks.\u003c/li\u003e\n\u003cli\u003eRecently, Diffusion Language Models (DLMs) have emerged as an alternative paradigm, generating text through iterative denoising instead of next-token prediction, thereby allowing parallel refinement of entire sequences.\u003c/li\u003e\n\u003cli\u003eWhile numerous diffusion-based architectures have been proposed, differences in evaluation protocols, datasets, inference budgets, and generation hyperparameters make it difficult to compare their functionalities and understand the trade-offs they offer.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19475v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Large Language Models (LLMs) have revolutionized language modeling through autoregressive generation, enabling strong performance across a wide range…\u003c/li\u003e\n\u003cli\u003eRecently, Diffusion Language Models (DLMs) have emerged as an alternative paradigm that generates text through iterative denoising rather than next-token predic…\u003c/li\u003e\n\u003cli\u003eWhile numerous diffusion-based architectures have been proposed, differences in evaluation protocols, datasets, inference budgets, and generation hyperparameter…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19494\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eHidden Anchors in Multi-Agent LLM Deliberation\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19494v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Multi-agent LLM deliberation (where agents exchange and revise answers over multiple rounds) is increasingly used to improve reasoning and accuracy, but how and why it works is rarely modeled.\u003c/li\u003e\n\u003cli\u003eThis deliberation mirrors how humans make decisions.\u003c/li\u003e\n\u003cli\u003eAs social animals, we are pulled both by the group—the herd mentality captured by classical models of opinion dynamics like those of DeGroot and Friedkin-Johnsen—and by our own intrinsic beliefs, which they do not capture.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003earXiv:2606.19494v1 Announce Type: new\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eAbstract: Multi-agent LLM deliberation, where agents exchange and revise answers over several rounds, is increasingly used to improve reasoning and accuracy, ye…\u003c/li\u003e\n\u003cli\u003eSuch deliberation mirrors how humans reach decisions\u003c/li\u003e\n\u003cli\u003eAs social animals we are pulled both by the group, the herd effect that classical opinion-dynamics models such as DeGroot and Friedkin\u0026ndash;Johnsen capture, and by…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19501\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eDeXposure-Claw: An Agentic System for DeFi Risk Supervision\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19501v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Decentralized finance exposes supervisors to fast-moving, networked credit risks.\u003c/li\u003e\n\u003cli\u003eGeneral-purpose LLM agents are ill-suited for this environment: they over-read weak evidence and recommend high-stakes interventions, while existing evaluations do not provide regulators with a consistent method to measure the resulting false positives.\u003c/li\u003e\n\u003cli\u003eWe introduce DeXposure-Claw, a forecast-grounded agentic supervision system that guides LLM decisions through structured evidence: (1) DeXposure-FM, a graph time-series foundation model that predicts future exposure networks; (2) Deterministic monitors and stress scenarios then translate these predictions into type alerts, attribution signals, and scenario evidence; (3) Data health and trust gates limit escalation before DeXposure-Claw issues auditable regulatory tickets with justifications.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN 要点:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19501v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Decentralized finance exposes supervisors to fast-moving, networked credit risks\u003c/li\u003e\n\u003cli\u003eGeneral-purpose LLM agents fit this setting poorly: they over-read weak evidence and recommend high-stakes interventions, while existing evaluations offer no re…\u003c/li\u003e\n\u003cli\u003eWe introduce DeXposure-Claw, a forecast-grounded agentic supervision system that routes LLM decisions through structured evidence: (1) DeXposure-FM, a graph tim…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19509\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eLLM Doesn\u0026rsquo;t Know What It Doesn\u0026rsquo;t Know: Detecting Epistemic Blind Spots via Cross-Model Attribution Divergence on Clinical Tabular Data\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19509v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Large Language Models (LLMs) are increasingly applied to structured clinical data, but whether they can recognize the limitations of their knowledge in such tasks remains to be explored.\u003c/li\u003e\n\u003cli\u003eWe investigate this issue through the lens of cross-model attribution divergence, aiming to reduce cognitive uncertainty in structured tasks by comparing Qwen 2.5 7B and XGBoost on prediction tasks via attribution divergence analysis.\u003c/li\u003e\n\u003cli\u003eFirstly, LLM linguistic confidence is epistemically vacuous, outputting near-constant values (0.856-0.937) regardless of whether accuracy is 49% or 75.3%, tracking prompt format rather than prediction quality.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN 要点:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19509v1 Announce Type: new\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eAbstract: Large language models (LLMs) are increasingly applied to structured clinical data, yet whether they can recognize the limits of their own knowledge on…\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eWe study this question through the lens of cross-model attribution divergence with the goal of reducing epistemic uncertainty for structured tasks, comparing Qw…\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eWe report four findings\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19522\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eREVEAL++: Differentiable Phenotypic Grouping for Vision-Language Retinal Modeling of Alzheimer\u0026rsquo;s Disease Risk\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublish Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19522v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: The retina offers a non-invasive window into neurodegenerative diseases, capturing subtle structural patterns associated with the risk of future cognitive decline.\u003c/li\u003e\n\u003cli\u003eVision-language alignment frameworks such as REVEAL have shown that pairing retinal fundus images with structured clinical risk narratives can improve early prediction of Alzheimer\u0026rsquo;s Disease (AD).\u003c/li\u003e\n\u003cli\u003eA key design choice in these methods is the use of phenotypic grouping, where individuals with similar risk profiles are treated as multi-positive pairs during contrastive learning.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19522v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: The retina offers a noninvasive window into neurodegenerative disease, capturing subtle structural patterns associated with a risk of future cognitive…\u003c/li\u003e\n\u003cli\u003eVision-language alignment frameworks such as REVEAL have shown that pairing retinal fundus images with structured clinical risk narratives improves early predic…\u003c/li\u003e\n\u003cli\u003eA key design choice in these approaches is the use of phenotypic grouping, where individuals with similar risk profiles are treated as multi-positive pairs duri…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19527\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eEmergent Alignment\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublish Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19527v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Can Large Language Models (LLMs) discern when their own outputs are misaligned with human ethics?\u003c/li\u003e\n\u003cli\u003eWe endow an LLM with a conscience step that reviews its own reasoning and outputs, and we extend the training loss with an alignment component using Direct Preference Optimization (DPO) to guide the model away from unethical outputs.\u003c/li\u003e\n\u003cli\u003eThe result is an online technique that can tune models across a wide range of applications: training, fine-tuning, adversarial prompting, and zero-shot learning.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19527v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Can Large Language Models (LLMs) discern when their own outputs are misaligned with human ethics\u003c/li\u003e\n\u003cli\u003eAnd can they self-correct\u003c/li\u003e\n\u003cli\u003eWe endow an LLM with a conscience step that reviews its own reasoning and outputs, and we extend the training loss with an alignment component using Direct Pref…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19538\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eITNet: A Learnable Integral Transform That Subsumes Convolution, Attention, and Recurrence\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19538v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Convolutional networks, recurrent networks, and transformers each encode different inductive biases—locality, sequential memory, and content-related pairwise interactions—and have remained mathematically distinct since their inception.\u003c/li\u003e\n\u003cli\u003eWe show that this fragmentation reflects not a fundamental diversity in how signals should be processed, but rather incomplete views of a single underlying mathematical object: the learnable integral transform.\u003c/li\u003e\n\u003cli\u003eWe introduce the Integral Transform Network (ITNet), a unified architecture built around a learnable kernel that jointly depends on position and features.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19538v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Convolutional networks, recurrent networks, and transformers each encode different inductive biases \u0026ndash; locality, sequential memory, and content-depend…\u003c/li\u003e\n\u003cli\u003eWe show that this fragmentation reflects not a fundamental diversity in how signals should be processed, but rather incomplete views of a single underlying math…\u003c/li\u003e\n\u003cli\u003eWe introduce the Integral Transform Network (ITNet), a unified architecture built around a learnable kernel that depends jointly on positions and features\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19559\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eUncertainty Decomposition for Clarification Seeking in LLM Agents\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19559v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Recent position papers argue that the classical aleatoric/epistemic uncertainty framework is insufficient for interactive large language model (LLM) agents and highlight the lack of canonical, decomposable, and communicable representations of uncertainty that could unlock new agent capabilities, such as actively seeking clarification and building shared mental models.\u003c/li\u003e\n\u003cli\u003ePractical deployment constraints—black-box APIs, interactive latency budgets, and the absence of labeled trajectories—rule out log-probability-based, multi-sampling, and training-based methods, making prompt-based estimation the most viable family of approaches for surfacing such signals at deployment time.\u003c/li\u003e\n\u003cli\u003eWe answer this call with a simple prompt-based decomposition that separates action confidence from request uncertainty (u), enabling the agent to ask for clarification when task specifications are ambiguous.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19559v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Recent position papers argue that the classical aleatoric/epistemic uncertainty framework is insufficient for interactive large language model (LLM) a…\u003c/li\u003e\n\u003cli\u003ePractical deployment constraints \u0026ndash; black-box APIs, interactive latency budgets, and the absence of labeled trajectories \u0026ndash; rule out logprob-based, multi-sampli…\u003c/li\u003e\n\u003cli\u003eWe answer this call with a simple prompt-based decomposition that separates action confidence from request uncertainty (u), enabling the agent to ask for clarif…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"arxiv-cscl-b_introsearch\"\u003e\n  ArXiv cs.CL (B_intro+search)\n  \u003ca class=\"heading-link\" href=\"#arxiv-cscl-b_introsearch\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19344\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eExposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregation\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19344v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Large Language Models (LLMs) exhibit representational and syntactic biases that are difficult to evaluate due to the stochastic nature of text generation.\u003c/li\u003e\n\u003cli\u003eStandard auditing methods rely on a single output inspection or static automated metrics.\u003c/li\u003e\n\u003cli\u003eThese approaches obscure the underlying probability distributions and fail to capture biases hidden in lower-probability generation branches.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19344v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Large Language Models (LLMs) exhibit representational and syntactic biases that are difficult to evaluate due to the stochastic nature of text generation.\u003c/li\u003e\n\u003cli\u003eStandard auditing methods rely on a single output inspection or static automated metrics\u003c/li\u003e\n\u003cli\u003eThese approaches obscure the underlying probability distributions and fail to capture biases hidden in lower-probability generation branches\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19345\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eEnsembles of Large Language Models for Identifying EQ-5D Studies in PubMed Based on Their Abstracts\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19345v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: The rapid increase in scientific publications has led to manual study screening in Systematic Literature Reviews (SLRs) becoming increasingly resource-intensive, inefficient, and inconsistent.\u003c/li\u003e\n\u003cli\u003eClassifying studies that clearly report health-related quality-of-life outcomes (e.g., EQ-5D data) requires a high level of clinical interpretation, posing challenges for human reviewers.\u003c/li\u003e\n\u003cli\u003eThis study investigates the use of Google\u0026rsquo;s Gemini and Gemma Large Language Models (LLMs) for automating EQ-5D detection in the PubMed biomedical database based solely on published abstracts.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19345v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: The rapid increase in scientific publications leads to the fact that manual study screening in systematic literature reviews (SLRs) is increasingly resource-intensive.\u003c/li\u003e\n\u003cli\u003eClassifying studies that clearly report health-related quality-of-life results, such as EQ-5D data, requires a high level of clinical interpretation and poses challenges for human reviewers.\u003c/li\u003e\n\u003cli\u003eThis study investigates the use of Google\u0026rsquo;s Gemini and Gemma large language models (LLMs) in automating EQ-5D detection in the PubMed biomedical database based solely on published abstracts.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19346\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eDisentangling Linguistic Relatedness from Task Alignment in Cross-Lingual Transfer\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19346v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: We investigate cross-lingual transfer by fine-tuning seven Large Language Models (4B\u0026ndash;671B parameters) in Arabic and evaluating zero-shot reading comprehension for Semitic languages and non-Semitic controls.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eIn dense and Mixture-of-Experts architectures, we find no evidence of Semitic-specific transfer: models with weak baselines improve significantly across all languages, while strong baseline models show only marginal gains, regardless of language family.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eA chain-of-thought ablation reinforces this finding — the same models that benefit most from fine-tuning benefit equally from inference-time reasoning, suggesting that both mechanisms address task format alignment rather than cross-lingual knowledge transfer.\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19346v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: We study cross-lingual transfer by fine-tuning seven large language models (4B\u0026ndash;671B parameters) on Arabic and evaluating zero-shot reading comprehens…\u003c/li\u003e\n\u003cli\u003eAcross dense and Mixture-of-Experts architectures, we find no evidence of Semitic-specific transfer: models with weak baselines improve dramatically across all…\u003c/li\u003e\n\u003cli\u003eA chain-of-thought ablation reinforces this finding \u0026ndash; the same models that benefit most from fine-tuning benefit equally from inference-time reasoning, suggest…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19347\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eHow LLMs Fail and Generalize in RTL Coding for Hardware Design?\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19347v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Translating sequential programming priors into the parallel temporal logic of hardware design remains a crucial bottleneck for Large Language Models (LLMs).\u003c/li\u003e\n\u003cli\u003eTo investigate this, we introduce a new error taxonomy grounded in problem solvability, inspired by cognitive theory.\u003c/li\u003e\n\u003cli\u003eOur taxonomy categorizes failures into syntactic, semantic, solvable functional, and unsolvable functional types.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19347v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Translating sequential programming priors into the parallel temporal logic of hardware design remains a crucial bottleneck for large language models(L…\u003c/li\u003e\n\u003cli\u003eTo investigate this, we introduce a new error taxonomy grounded in problem solvability, inspired by cognitive theory\u003c/li\u003e\n\u003cli\u003eOur taxonomy categorizes failures into syntactic, semantic, solvable functional, and unsolvable functional types\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19348\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eDeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19348v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: We present a preview of the DeepSeek-V4 series, featuring two powerful Mixture-of-Experts (MoE) language models — DeepSeek-V4-Pro with 1.6T parameters (49B active) and DeepSeek-V4-Flash with 284B parameters (13B active) — both supporting a 1-million-token context length.\u003c/li\u003e\n\u003cli\u003eThe DeepSeek-V4 series incorporates several key architectural and optimization upgrades: (1) a hybrid attention architecture combining Compressed Sparse Attention (CSA) and Heavy Compressed Attention (HCA) for improved long-context efficiency; (2) manifold-constrained hyper-connections (mHC) to enhance traditional residual connections; and (3) the Muon optimizer for faster convergence and higher training stability.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eWe pre-train both models on more than 32T diverse and high-quality tokens, followed by a comprehensive post-training pipeline that unlocks and further enhances their capabilities.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19348v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models \u0026ndash; DeepSeek-V4-Pro with 1.6T paramet…\u003c/li\u003e\n\u003cli\u003eDeepSeek-V4 series incorporate several key upgrades in architecture and optimization: (1) a hybrid attention architecture that combines Compressed Sparse Attent…\u003c/li\u003e\n\u003cli\u003eWe pre-train both models on more than 32T diverse and high-quality tokens, followed by a comprehensive post-training pipeline that unlocks and further enhances…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19349\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eWhere to Place the Query? Unveiling and Mitigating Positional Bias in In-Context Learning for Diffusion LLMs via Decoding Dynamics\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19349v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: While In-Context Learning (ICL) is extensively studied in Autoregressive (AR) LLMs, its mechanism within Diffusion Large Language Models (dLLMs) remains largely unexplored.\u003c/li\u003e\n\u003cli\u003eUnlike AR models restricted by unidirectional causal masking, dLLMs intrinsically utilize bidirectional attention, offering extensive spatial flexibility for query placement.\u003c/li\u003e\n\u003cli\u003eUnfortunately, current practices conventionally inherit AR-style trailing-query templates, often overlooking the structural paradigm shift.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19349v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: While In-Context Learning (ICL) is extensively studied in Autoregressive (AR) LLMs, its mechanism within Diffusion Large Language Models (dLLMs) remai…\u003c/li\u003e\n\u003cli\u003eUnlike AR models restricted by unidirectional causal masking, dLLMs intrinsically utilize bidirectional attention, offering extensive spatial flexibility for qu…\u003c/li\u003e\n\u003cli\u003eUnfortunately, current practices conventionally inherit AR-style trailing-query templates, often overlooking the structural paradigm shift\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19350\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003ePruning via Causal Attribution Preserves Reasoning Performance in Large Language Models\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19350v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Large Language Models (LLMs) excel at multi-step reasoning but incur significant inference costs.\u003c/li\u003e\n\u003cli\u003eWe introduce Causal Attribution Pruning (CAP), a training-free method that identifies critical attention heads by measuring their causal impact on reasoning tasks, and uses these head-level scores to guide fine-grained weight pruning.\u003c/li\u003e\n\u003cli\u003eFor each attention head, CAP estimates the expected performance drop when that head is masked during a forward pass on a small set of reasoning problems.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19350v1 Announce Type: new\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eAbstract: Large language models (LLMs) excel at multi-step reasoning but incur substantial inference cost\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eWe introduce Causal Attribution Pruning (CAP), a training-free method that identifies critical attention heads by measuring their causal impact on reasoning tas…\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eFor each attention head, CAP estimates the expected performance degradation when the head is masked during forward passes on a small calibration set of reasonin…\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19351\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eDetecting Hallucinations for Large Language Model-based Knowledge Graph Reasoning\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19351v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Knowledge graph (KG) reasoning infers new knowledge from existing facts and is widely applied in question answering, recommendation, and decision support.\u003c/li\u003e\n\u003cli\u003eWith the rapid development of large language models (LLMs), LLM-based KG reasoning frameworks have become increasingly popular by leveraging retrieved KG information.\u003c/li\u003e\n\u003cli\u003eHowever, hallucinations in LLMs remain a critical issue.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19351v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Knowledge graph (KG) reasoning infers new knowledge from existing facts and is widely applied in question answering, recommendation, and decision supp…\u003c/li\u003e\n\u003cli\u003eWith the rapid development of large language models (LLMs), LLM-based KG reasoning frameworks have become increasingly popular by leveraging retrieved KG inform…\u003c/li\u003e\n\u003cli\u003eHowever, hallucinations in LLMs remain a critical issue\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19352\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eSign-Language Datasets at Scale: A Comprehensive Survey on Resources, Benchmarks, and Annotation Standards\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19352v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Sign languages are expressive visual languages used by Deaf and Hard-of-Hearing (DHH) communities.\u003c/li\u003e\n\u003cli\u003eDespite substantial progress in sign-language recognition, translation, and production, advances remain constrained by fragmented datasets, inconsistent annotations, and limited language coverage.\u003c/li\u003e\n\u003cli\u003eExisting benchmarks often fail to reflect real-world communication needs, and a systematic analysis of these limitations remains limited.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19352v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Sign languages are expressive visual languages used by Deaf and Hard-of-Hearing (DHH) communities\u003c/li\u003e\n\u003cli\u003eDespite substantial progress in sign-language recognition, translation, and production, advances remain constrained by fragmented datasets, inconsistent annotat…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eExisting benchmarks often fail to reflect real-world communication needs, and systematic analyses of these limitations remain limited\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19353\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eQuantifying Aleatoric Uncertainty of In-Context Learning for Robust Measure of LLM Prediction Confidence\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19353v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: In-Context Learning (ICL) allows LLMs to adapt to new tasks from a few demonstrations, but its reliability remains a concern: predictions are highly sensitive to both prompt design and the model\u0026rsquo;s ability to understand context, blurring whether failures are caused by data properties or model limitations.\u003c/li\u003e\n\u003cli\u003eUncertainty decomposition (separating aleatoric from epistemic sources) is particularly crucial in this setting, yet existing methods designed for standard generative tasks cannot capture the unique dynamics of ICL.\u003c/li\u003e\n\u003cli\u003eTo address this, we introduce a concept of self-function vectors, built upon Bayesian views and the mechanistic interpretability of ICL.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19353v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: In-Context Learning (ICL) allows LLMs to adapt to new tasks from a few demonstrations, but its reliability remains a concern: predictions are highly s…\u003c/li\u003e\n\u003cli\u003eUncertainty decomposition-separating aleatoric from epistemic sources-is particularly crucial in this setting, yet existing methods, designed for standard gener…\u003c/li\u003e\n\u003cli\u003eTo address this, we introduce a concept of self-function vectors, built upon Bayesian views and the mechanistic interpretability of ICL\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"arxiv-cslg-b_introsearch\"\u003e\n  ArXiv cs.LG (B_intro+search)\n  \u003ca class=\"heading-link\" href=\"#arxiv-cslg-b_introsearch\"\u003e\n    \u003ci class=\"fa-solid fa-link\" aria-hidden=\"true\" title=\"Link to heading\"\u003e\u003c/i\u003e\n    \u003cspan class=\"sr-only\"\u003eLink to heading\u003c/span\u003e\n  \u003c/a\u003e\n\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19361\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eComputational Identifiability\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19361v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Identification conditions describe the computability of a target query or parameter of interest as a function of the type and amount of available information.\u003c/li\u003e\n\u003cli\u003eIn causal identification, this information is often expressed in the form of a causal graph, and data are observed or collected for some subset of variables in the graph.\u003c/li\u003e\n\u003cli\u003eTarget queries may be for a single effect alone or for a class of effects in a given model.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19361v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Identification conditions describe the computability of a target query or parameter of interest as a function of the type and amount of information av…\u003c/li\u003e\n\u003cli\u003eIn causal identification, this information is often expressed in the form of a causal graph, and data are observed or collected for some subset of variables in…\u003c/li\u003e\n\u003cli\u003eTarget queries may be for a single effect alone or for a class of effects in a given model\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19363\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eWhen to Trust, How to Distill: Multi-Foundation Model Guidance for Lightweight, Robust Scientific Time Series Forecasting\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19363v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: The deployment of Time-Series Foundation Models (TSFMs) in physical sciences is hindered by a critical trade-off: while these models encode rich, universal temporal dynamics, they suffer from severe distributional misalignment when applied zero-shot to specific scientific domains, and their computational cost prevents deployment in edge computing sensor networks.\u003c/li\u003e\n\u003cli\u003eWe address a fundamental challenge: How can we extract latent structural knowledge from misaligned foundation models (FMs) to train lightweight, specialized forecasters?\u003c/li\u003e\n\u003cli\u003eWe propose Gated Uncertainty-Aware Routing for Distillation (Guard), a novel framework that reframes multi-teacher distillation as an instance-wise decision process with two adaptive mechanisms: (1) a contextual router that dynamically selects the most relevant teacher based on local input statistics, leveraging the complementarity between different foundation models; and (2) an uncertainty-gated temperature mechanism that acts as a \u0026ldquo;circuit breaker,\u0026rdquo; automatically weakening the distillation intensity when teacher confidence deviates from domain reality.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19363v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: The deployment of Time-Series Foundation Models (TSFMs) in physical sciences is hindered by a critical trade-off: while these models encode rich, univ…\u003c/li\u003e\n\u003cli\u003eWe address a fundamental challenge: How can we extract latent structural knowledge from misaligned foundation models (FM) to train lightweight, specialized fore…\u003c/li\u003e\n\u003cli\u003eWe propose Gated Uncertainty-Aware Routing for Distillation (Guard), a novel framework that reframes multiteacher distillation as an instance-wise decision proc…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19364\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eClosing the Social-Semantic Gap: SPSD for Edge-Based Prompt Compression in Cloud LLM Inference\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublished: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19364v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: The prefill stage of Large Language Model (LLM) inference is a growing contributor to cloud-scale energy costs.\u003c/li\u003e\n\u003cli\u003eMany consumer-support and conversational prompts contain social scaffolding: politeness markers, apologetic preambles, repetition, and rapport-building language that are important for human communication but contain little marginal information for machine reasoning.\u003c/li\u003e\n\u003cli\u003eWe call this discrepancy the Social-Semantic Gap.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19364v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: The prefill stage of Large Language Model (LLM) inference is a growing contributor to cloud-scale energy cost\u003c/li\u003e\n\u003cli\u003eMany consumer-support and conversational prompts contain social scaffolding: politeness markers, apologetic preamble, repetition, and rapport-building language…\u003c/li\u003e\n\u003cli\u003eWe call this discrepancy the Social-Semantic Gap\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19365\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003ePerformance Analysis and Optimization of 3D Generative Diffusion Models across GPU Architectures\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19365v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Diffusion models are crucial for high-fidelity 3D MRI synthesis, yet their deployment remains constrained by the substantial GPU resource demands arising from hundreds of U-Net evaluations per sample and highly heterogeneous kernel behavior.\u003c/li\u003e\n\u003cli\u003eThis paper performs a comprehensive performance analysis of the state-of-the-art medical diffusion model, Med-DDPM, across three generations of NVIDIA architectures to study kernel-level runtime failures, instruction mix characteristics, memory system utilization, warp-level activity, and profiler priority score estimation.\u003c/li\u003e\n\u003cli\u003eWe show that training is overwhelmingly dominated by cuDNN convolution and implicit-GEMM kernels, with inefficiencies arising from memory-access patterns, tensor layout transformations, and limited tensor core utilization.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19365v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Diffusion models have become essential for high-fidelity 3D MRI synthesis, yet their deployment remains constrained by substantial GPU resource demand…\u003c/li\u003e\n\u003cli\u003eThis paper performs a comprehensive performance analysis of the state-of-the-art medical diffusion model, Med-DDPM, across three generations of NVIDIA architect…\u003c/li\u003e\n\u003cli\u003eWe show that training is overwhelmingly dominated by cuDNN convolution and implicit-GEMM kernels, with inefficiencies arising from memory-access patterns, tenso…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19366\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eInformation Lattice Learning as Probabilistic Graphical Model Structure Learning\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19366v1 Announcement Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Information lattice learning (ILL) learns interpretable rules of a signal by alternately projecting the signal onto a partition lattice that encodes a hierarchy of abstractions and lifting selected rules back to the signal domain.\u003c/li\u003e\n\u003cli\u003eWhen the signal is a probability mass function, we show the probabilistic rules learned by ILL admit a natural probabilistic graphical model (PGM) interpretation and develop this interpretation in detail.\u003c/li\u003e\n\u003cli\u003eA partition in ILL induces a deterministic quotient variable, and a rule is the marginal law of that quotient variable.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Key Points:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19366v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Information lattice learning (ILL) learns interpretable rules of a signal by alternately projecting the signal onto a partition lattice that encodes a…\u003c/li\u003e\n\u003cli\u003eWhen the signal is a probability mass function, we show the probabilistic rules learned by ILL admit a natural probabilistic graphical model (PGM) interpretatio…\u003c/li\u003e\n\u003cli\u003eA partition in ILL induces a deterministic quotient variable, and a rule is the marginal law of that quotient variable\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19367\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eWeibull Weight-Scale Parameter Evolution under AdamW Training Dynamics\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003ePublication Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19367v1 Announcement Type: new.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eAbstract: Building on a two-parameter Weibull framework for diagnosing transformer weight distributions, we study why the Weibull weight-scale parameter $\\lambda$ grows, overshoots, and then relaxes during AdamW training.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eWe derive a leading-order three-force decomposition of the squared weight norm from the AdamW update: an alignment force measuring the correlation between weights and the adaptive update direction, an injection force from the adaptive step size, and a decay force from decoupled weight decay.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eOn self-trained Pythia-70M models with ground-truth optimizer moments, alignment dominates the rise phase, contributing 88-94% of the absolute force budget across four random seeds, and is robust to overweight removal.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eEN Highlights:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19367v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Building on a two-parameter Weibull framework for diagnosing transformer weight distributions, we study why the Weibull weight-scale parameter $\\lambd…\u003c/li\u003e\n\u003cli\u003eWe derive a leading-order three-force decomposition of the squared weight norm from the AdamW update: an alignment force measuring the correlation between weigh…\u003c/li\u003e\n\u003cli\u003eOn self-trained Pythia-70M models with ground-truth optimizer moments, alignment dominates the rise phase, contributing 88-94% of the absolute force budget acro…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19369\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eZero-Inflated Gaussian Distributions Enable Parameter-Space Sparsity in Estimation-of-Distribution Algorithms\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eRelease Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19369v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Estimation-of-distribution algorithms (EDAs) are a powerful class of evolutionary methods for black-box optimization, especially when little is known about the objective\u0026rsquo;s structure.\u003c/li\u003e\n\u003cli\u003eWhereas classical evolutionary algorithms rely on hand-designed mutation and crossover operators, which are hard to devise for unknown problem structures and a source of bias, EDAs completely sidestep operator design: they fit a probability distribution to the best individuals and sample the next generation from it.\u003c/li\u003e\n\u003cli\u003eEDAs are well established on continuous parameter spaces, but they have not previously been generalized to sparse ones, in which most coefficients of a good solution are exactly zero.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003eEN Highlights:\n\u003cul\u003e\n\u003cli\u003earXiv:2606.19369v1 Announce Type: new\u003c/li\u003e\n\u003cli\u003eAbstract: Estimation-of-distribution algorithms (EDAs) are a powerful class of evolutionary methods for black-box optimization, especially when little is known…\u003c/li\u003e\n\u003cli\u003eWhereas classical evolutionary algorithms rely on hand-designed mutation and crossover operators, hard to devise for unknown problem structures, and a source of…\u003c/li\u003e\n\u003cli\u003eEDAs are well established on continuous parameter spaces, but they have not previously been generalized to sparse ones, in which most coefficients of a good sol…\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19370\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eHuman-like autonomy emerges from self-play and a pinch of human data\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eRelease Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19370v1 Announce Type: new.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eAbstract: Self-play reinforcement learning has recently emerged as a way to train driving policies without any human data.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eIt uses cheap, large-scale simulations to substitute expensive, large-scale human driving demonstrations.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eA key limitation of this approach is that policies trained through pure self-play can learn effective but alien driving conventions incompatible with people.\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19371\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003eProMUSE: Progressive Multi-modal Uncertainty-guided Staged Evidential Alzheimer Disease Classification\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eRelease Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19371v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Alzheimer\u0026rsquo;s disease (AD) is a fatal disorder that destroys memory and cognitive skills in the elderly population.\u003c/li\u003e\n\u003cli\u003eMost treatments for AD are effective in the early stage, leading to an increasing demand for early AD diagnosis.\u003c/li\u003e\n\u003cli\u003eAD diagnosis increasingly relies on multimodal data such as clinical assessments, structural Magnetic Resonance Imaging (MRI), and Positron Emission Tomography (PET) imaging.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"https://arxiv.org/abs/2606.19373\"  class=\"external-link\" target=\"_blank\" rel=\"noopener\"\u003ecAPM: Continual AI-Assisted Pace-Mapping with Active Learning\u003c/a\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eRelease Time: 2026-06-19 12:00 Beijing Time\u003c/li\u003e\n\u003cli\u003eAbstract: - arXiv:2606.19373v1 Announce Type: new.\n\u003cul\u003e\n\u003cli\u003eAbstract: Ventricular tachycardia is a life-threatening rhythm disorder and a major cause of sudden cardiac death.\u003c/li\u003e\n\u003cli\u003ePace-mapping is a clinical procedure for identifying the intervention target during catheter ablation of VT.\u003c/li\u003e\n\u003cli\u003eIt requires clinicians to pace different sites of the ventricle and quickly interpret the resulting electrocardiograms to determine where to pace next or if a target site has been identified.\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003eIt requires clinicians to pace different sites in the ventricles and rapidly interpret the resulting electrocardiograms to determine where to pace next or wheth…\u003c/p\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n",
  "wordCount": 7804,
  "readingTime": 37,
  "tableOfContents": "\u003cnav id=\"TableOfContents\"\u003e\n  \u003cul\u003e\n    \u003cli\u003e\u003ca href=\"#-in-depth-guide-to-this-issues-watch-list\"\u003e📖 In-depth Guide to This Issue\u0026rsquo;s Watch List\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#-ai-hot-topics-on-x\"\u003e🌐 AI Hot Topics on X\u003c/a\u003e\n      \u003cul\u003e\n        \u003cli\u003e\u003ca href=\"#topic-1-transformer-pioneer-noam-shazeer-leaves-google-for-openai\"\u003eTopic 1: Transformer Pioneer Noam Shazeer Leaves Google for OpenAI\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#topic-2-loop-engineering-turns-ai-agents-into-self-sustaining-coders\"\u003eTopic 2: Loop Engineering Turns AI Agents into Self-Sustaining Coders\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#topic-3-anthropic-fixes-claude-code-usage-bug-for-premium-users\"\u003eTopic 3: Anthropic Fixes Claude Code Usage Bug for Premium Users\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#topic-4-zais-glm-52-tops-open-ai-model-charts-with-strong-benchmarks\"\u003eTopic 4: Z.ai\u0026rsquo;s GLM-5.2 Tops Open AI Model Charts with Strong Benchmarks\u003c/a\u003e\u003c/li\u003e\n      \u003c/ul\u003e\n    \u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#-influencer-insights\"\u003e💡 Influencer Insights\u003c/a\u003e\u003c/li\u003e\n  \u003c/ul\u003e\n\n  \u003cul\u003e\n    \u003cli\u003e\u003ca href=\"#1-todays-focus-the-rise-of-on-device-models-and-their-engineering-practices\"\u003e1. Today\u0026rsquo;s Focus: The Rise of On-device Models and Their Engineering Practices\u003c/a\u003e\n      \u003cul\u003e\n        \u003cli\u003e\u003ca href=\"#11-on-device-model-capabilities-validated-with-breakthroughs-in-both-performance-and-quality\"\u003e1.1 On-device Model Capabilities Validated, with Breakthroughs in Both Performance and Quality\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#12-ai-coding-tools-enter-a-clash-of-titans-and-a-new-stage-of-automation\"\u003e1.2 AI Coding Tools Enter a \u0026ldquo;Clash of Titans\u0026rdquo; and a New Stage of Automation\u003c/a\u003e\u003c/li\u003e\n      \u003c/ul\u003e\n    \u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#2-noteworthy-unique-perspectives-and-industry-foresight\"\u003e2. Noteworthy Unique Perspectives and Industry Foresight\u003c/a\u003e\n      \u003cul\u003e\n        \u003cli\u003e\u003ca href=\"#21-the-technological-paradigm-shift-from-correlation-to-causality\"\u003e2.1 The Technological Paradigm Shift from \u0026ldquo;Correlation\u0026rdquo; to \u0026ldquo;Causality\u0026rdquo;\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#22-the-evolution-of-vibe-coding-from-requirements-first-to-contract-first\"\u003e2.2 The Evolution of Vibe Coding: From \u0026ldquo;Requirements-First\u0026rdquo; to \u0026ldquo;Contract-First\u0026rdquo;\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#23-personal-development-philosophy-in-the-ai-era\"\u003e2.3 Personal Development Philosophy in the AI Era\u003c/a\u003e\u003c/li\u003e\n      \u003c/ul\u003e\n    \u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#3-recommended-tools--resources\"\u003e3. Recommended Tools \u0026amp; Resources\u003c/a\u003e\n      \u003cul\u003e\n        \u003cli\u003e\u003ca href=\"#31-development--design-tools\"\u003e3.1 Development \u0026amp; Design Tools\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#32-productivity--automation\"\u003e3.2 Productivity \u0026amp; Automation\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#33-cutting-edge-models--frameworks\"\u003e3.3 Cutting-Edge Models \u0026amp; Frameworks\u003c/a\u003e\u003c/li\u003e\n      \u003c/ul\u003e\n    \u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#-appendix-todays-watch-list-source-updates\"\u003e📚 Appendix: Today\u0026rsquo;s Watch List Source Updates\u003c/a\u003e\n      \u003cul\u003e\n        \u003cli\u003e\u003ca href=\"#y-combinator-podcast-b_introsearch\"\u003eY Combinator Podcast (B_intro+search)\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#all-in-podcast-a_full\"\u003eAll-In Podcast (A_full)\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#stratechery-by-ben-thompson-a_full\"\u003eStratechery by Ben Thompson (A_full)\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#two-minute-papers-b_introsearch\"\u003eTwo Minute Papers (B_intro+search)\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#arxiv-csai-b_introsearch\"\u003eArXiv cs.AI (B_intro+search)\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#arxiv-cscl-b_introsearch\"\u003eArXiv cs.CL (B_intro+search)\u003c/a\u003e\u003c/li\u003e\n        \u003cli\u003e\u003ca href=\"#arxiv-cslg-b_introsearch\"\u003eArXiv cs.LG (B_intro+search)\u003c/a\u003e\u003c/li\u003e\n      \u003c/ul\u003e\n    \u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/nav\u003e",
  "isDraft": false
}