2026-10-08 AI日更 | ChatGPT 扩至青少年规划与酒店预订，DeepMind AlphaGenome Atlas 引领生命科学可校准 链接到标题

今日核心：智能体进入真实消费工作流——ChatGPT for Teens 上线 College Planner、闪卡与测验，Radisson 酒店预订接入 ChatGPT；Stratechery 拆解 Apple/LG 智能家居与 Amazon 智能体标准。生命科学方向，DeepMind AlphaGenome Atlas 与可校准随机试验模型突显可追溯、可校准趋势。可解释性 arXiv 密集更新。

📖 本期 Watch List 深度导读 链接到标题

今天值得重点读的有三条线。一是智能体从演示走向真实工作流：OpenAI 为青少年推出 College Planner、闪卡与测验，Radisson 联合 Accenture 把酒店发现与预订接入 ChatGPT，Stratechery 对 Apple/LG 智能家居和 Amazon 智能体标准的拆解也值得产品与生态团队关注。二是可解释性出现机制化密集更新：Transformer 拒绝机制的稀疏性、潜在结构可操作性、RoPE 剪枝与持续学习稳定化等 arXiv 工作集中出现，适合对齐与基础设施团队跟进。三是生命科学与专业证据方向：热传的 DeepMind 更新指向 AlphaGenome Atlas，另有 4B 开放模型对随机试验给出校准概率、生物医学翻译压缩等，说明“可追溯、可校准”正成为领域模型的新指标。

🌐 X 平台 AI 热点快讯 链接到标题

话题 1:Grok Bot Routes Tasks to Top AI Models for Best Results 链接到标题

• 分类:AI · News
• 概况:热度时间:2 days ago,相关帖子数:64000
• 是什么事:Grok 机器人开始将用户任务自动路由到多个顶尖 AI 模型，以返回最佳结果。
• 为什么重要:这表明 AI 助手正从单一模型能力竞争转向多模型调度与优化，可能提升回答质量并重塑模型生态。
• 讨论概况:X 上的讨论焦点包括路由机制是否真能选出最优模型、响应速度与成本，以及与 ChatGPT、Claude 等产品的实际差距。

话题 2:Gemini 4 Argon Appears in Google Services Ahead of Launch 链接到标题

• 分类:AI · News
• 概况:热度时间:1 day ago,相关帖子数:3500
• 是什么事:谷歌尚未发布的 Gemini 4 Argon 模型出现在 Google 服务中，引发对其即将发布的关注。
• 为什么重要:这被视为谷歌下一代旗舰模型发布前的信号，可能影响 AI 模型竞争格局和行业基准评估。
• 讨论概况:X 上围绕其基准测试表现存在分歧：该模型在 19 项测试中赢下 13 项，但输掉项目的差距更大，讨论焦点在于胜利数量与实际领先幅度是否匹配。

话题 3:OpenAI Releases 722 AI-Generated Math Breakthroughs on GitHub 链接到标题

• 分类:AI · News
• 概况:热度时间:1 day ago,相关帖子数:88000
• 是什么事:OpenAI 在 GitHub 上发布了 722 个由 AI 生成的数学突破性成果。
• 为什么重要:这表明 AI 在数学推理和科学发现方面的能力进一步提升，可能加速研究进程并改变科研方式。
• 讨论概况:X 上的讨论焦点包括这些成果是否经过严格验证、能否被视为真正的数学突破，以及 AI 生成内容对学术研究和科研人员的冲击。

话题 4:Germany Pushes for EU Approval of Tesla’s Full Self-Driving Supervised 链接到标题

• 分类:AI · News
• 概况:热度时间:17 hours ago,相关帖子数:20000
• 是什么事:德国推动欧盟批准特斯拉的受监督全自动驾驶系统（FSD Supervised）。
• 为什么重要:若获批，特斯拉的驾驶辅助技术将首次进入欧盟核心市场，可能加速欧洲自动驾驶法规与竞争格局的演变。
• 讨论概况:X 上关注焦点包括德国监管态度对欧盟其他成员国的影响、特斯拉 FSD 与欧洲现有安全标准的兼容性，以及此举是否会给本地汽车制造商带来压力。

话题 5:Ben Affleck Reveals Python Coding and Neural Network Skills from VFX Work 链接到标题

• 分类:AI · Entertainment
• 概况:热度时间:1 day ago,相关帖子数:16000
• 是什么事:本·阿弗莱克透露自己在从事视觉特效工作期间掌握了 Python 编程和神经网络相关技能。
• 为什么重要:好莱坞一线演员公开谈论编程与神经网络，可能推动公众对 AI 在影视特效和内容制作中应用的关注。
• 讨论概况:X 上的讨论集中在对名人技术能力的惊讶与质疑、明星参与 AI 与影视制作的意义，以及编程和神经网络是否正成为创意行业的重要工具。

今日 X 上的 AI 舆情小结 链接到标题

今天舆论主线是 AI 正从单一模型能力竞争转向多模型调度、下一代模型卡位与垂直行业加速渗透并行的阶段，Grok 的多模型路由、谷歌 Gemini 4 Argon 的提前曝光和 OpenAI 的数学成果都集中体现了这一趋势。共识在于业界普遍认可多模型协同和更先进的推理能力会提升回答质量、推动科研与创意行业效率，但分歧也很明显：Grok 能否真正选出最优模型且兼顾成本速度、Gemini 胜项多但落后幅度大是否算领先、OpenAI 的 722 项数学成果能否经受严格验证，以及特斯拉 FSD 是否符合欧洲安全预期都存在争议。潜在风险集中在未经验证的 AI 科研成果可能冲击学术信任、自动驾驶监管放宽带来安全与本地产业压力，以及模型调度成本和名人技术光环可能加剧生态依赖或公众误解。

💡 大佬观点(Influencer Insights) 链接到标题

今日大佬观点暂缺,推荐阅读 Watch List 深度内容。

📚 附录:今日 Watch List 更新源列表 链接到标题

时间窗口:最近 3 天;覆盖 22 个源;共 35 条更新

Stratechery by Ben Thompson (A_full) 链接到标题

• Apple and LG, The House For Everyone Else, Agent Standards and Amazon
  • 发布时间:2026-10-07 18:01 北京时间
  • 摘要:【待翻译】- Apple is taking a smarter approach to the home than I expected, leaning into integration (with partners); then, what Amazon should do about agents.
    • $15 / month or $150 / year.
    • Substantial analysis of the news of the day delivered via three weekly emails or podcasts.
    • Stratechery Interviews.
    • Interviews with leading public CEOs, private company founders, and discussions with fellow analysts.
  • EN 要点:
    • Apple is taking a smarter approach to the home than I expected, leaning into integration (with partners); then, what Amazon should do about agents.

OpenAI Blog (A_full) 链接到标题

• Helping teens learn, plan, and shape the future of AI

  • 发布时间:2026-10-07 20:00 北京时间
  • 摘要:【待翻译】- College Planner is coming to ChatGPT for Teens to help students manage college applications, alongside new flashcards, quizzes, and a teen AI council.
    • This piece from OpenAI Blog explains how Helping teens learn, plan, and shape the future of AI shapes the broader AI and infrastructure landscape.
    • It also surfaces practical implications for founders, operators, and investors following Helping teens learn, plan, and shape the future of AI.
  • EN 要点:
    • College Planner is coming to ChatGPT for Teens to help students manage college applications, alongside new flashcards, quizzes, and a teen AI council.

• Radisson Hotel Group brings hotel discovery into ChatGPT

  • 发布时间:2026-10-07 15:00 北京时间
  • 摘要:【待翻译】- Radisson partnered with Accenture to build a ChatGPT plugin using OpenAI technology, helping travelers find, compare, and book hotels while planning their trips.
    • Radisson partnered with Accenture to build a ChatGPT plugin using OpenAI technology, helping travelers find, compare, and book hotels while planning their trips….
    • Radisson Hotel Group brings hotel discovery into ChatGPT.
  • EN 要点:
    • Radisson partnered with Accenture to build a ChatGPT plugin using OpenAI technology, helping travelers find, compare, and book hotels while planning their trips…

• GPT-6 and Intelligent UI for everyone

  • 发布时间:2026-10-07 08:00 北京时间
  • 摘要:【待翻译】- GPT‑6 is rolling out globally in ChatGPT with Intelligent UI, delivering faster responses with visuals and interactive experiences you can explore and use direc….
    • This piece from OpenAI Blog explains how GPT-6 and Intelligent UI for everyone shapes the broader AI and infrastructure landscape.
    • It also surfaces practical implications for founders, operators, and investors following GPT-6 and Intelligent UI for everyone.
  • EN 要点:
    • GPT‑6 is rolling out globally in ChatGPT with Intelligent UI, delivering faster responses with visuals and interactive experiences you can explore and use direc…

Two Minute Papers (B_intro+search) 链接到标题

• DeepMind’s New AI Just Cracked The Code Of Life
  • 发布时间:2026-10-07 21:29 北京时间
  • 摘要:【待翻译】- ❤️ Check out Lambda here and sign up for their GPU Cloud:.
    • 📝 AlphaGenome Atlas is available here:.
    • Adam Bridges, B Shang, Carlos Galarza, Christian Ahlin, Eric Tyson, Juan Benet, Lukas Biewald, Michael Tedder, Owen Skarpness, Ryan Stankye, Shawn Becker, Steef, Taras Bobrovytsky, Tazaur Sagenclaw, Tybie Fitzhugh, Ueli Gallizzi.
    • DeepMind’s New AI Just Cracked The Code Of Life.
  • EN 要点:
    • ❤️ Check out Lambda here and sign up for their GPU Cloud:
    • Thank you Google DeepMind for the invite
    • 📝 AlphaGenome Atlas is available here:
    • 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:

ArXiv cs.AI (B_intro+search) 链接到标题

• Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03872v1 Announce Type: new.
    • Abstract: AI agents are becoming increasingly capable of generating scientific code, but generating code is not the same as improving the algorithms behind it.
    • For numerical solvers, execution feedback can expose poor performance, but rarely reveals its underlying cause and how to address it.
    • We introduce Auto-Diagnosis and Skill Discovery (ADSD), a framework that links numerical diagnosis to reusable solver self-improvement.
  • EN 要点:
    • arXiv:2610.03872v1 Announce Type: new
    • Abstract: AI agents are becoming increasingly capable of generating scientific code, but generating code is not the same as improving the algorithms behind it
    • For numerical solvers, execution feedback can expose poor performance, but rarely reveals its underlying cause and how to address it
    • We introduce Auto-Diagnosis and Skill Discovery (ADSD), a framework that links numerical diagnosis to reusable solver self-improvement

• REACT: Physically and Chemically Consistent Reconstruction of Marine Active Tracers

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03888v1 Announce Type: new.
    • Abstract: Reconstructing global sea surface pH from sparse observations is critical for monitoring ocean acidification and understanding marine carbon cycling.
    • Traditional assimilation and inverse models are physically grounded but costly for large-scale reconstruction.
    • Recent black-box and physics-guided AI models improve efficiency, but are mainly designed for passive tracers, where the reconstructed variable is also the transported inventory.
  • EN 要点:
    • arXiv:2610.03888v1 Announce Type: new
    • Abstract: Reconstructing global sea surface pH from sparse observations is critical for monitoring ocean acidification and understanding marine carbon cycling
    • Traditional assimilation and inverse models are physically grounded but costly for large-scale reconstruction
    • Recent black-box and physics-guided AI models improve efficiency, but are mainly designed for passive tracers, where the reconstructed variable is also the tran…

• Proxy Confidence: Auditing Black-Box LLM Agents with a Surrogate’s Log-Probabilities

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03894v1 Announce Type: new.
    • Abstract: A deployed LLM agent emits tool calls, queries, and code that can be silently wrong – by the time the error surfaces, the action has run.
    • Frontier chat APIs hide the model’s token probabilities; the agent’s stated confidence barely beats chance on the mistakes that matter; and resampling does not help, since frontier models are highly repetitive, reproducing the same call across samples.
    • We recover the missing signal from a low-cost open-weight surrogate run in parallel.
  • EN 要点:
    • arXiv:2610.03894v1 Announce Type: new
    • Abstract: A deployed LLM agent emits tool calls, queries, and code that can be silently wrong – by the time the error surfaces, the action has run
    • Frontier chat APIs hide the model’s token probabilities; the agent’s stated confidence barely beats chance on the mistakes that matter; and resampling does not…
    • We recover the missing signal from a low-cost open-weight surrogate run in parallel

• SGAnalog: An End-to-End Circuit Benchmark from Open-Source Silicon Tapeouts

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03934v1 Announce Type: new.
    • Abstract: Existing analog integrated circuit design benchmarks make two questions hard to answer: whether a model has learned transferable circuit skills rather than recalled familiar examples, and whether its output works under defined process and test conditions.
    • We introduce a benchmark built from human-designed, open-source circuits associated with Tiny Tapeout manufacturing shuttles.
    • The collection contains 273 topologically distinct top-level designs.
  • EN 要点:
    • arXiv:2610.03934v1 Announce Type: new
    • Abstract: Existing analog integrated circuit design benchmarks make two questions hard to answer: whether a model has learned transferable circuit skills rather…
    • We introduce a benchmark built from human-designed, open-source circuits associated with Tiny Tapeout manufacturing shuttles
    • The collection contains 273 topologically distinct top-level designs

• MLLMs Fail to Refuse when Using Tools Agentically

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03938v1 Announce Type: new.
    • Abstract: Agentic multimodal large language models (MLLMs) have recently pushed the frontier of visual reasoning by calling tools such as zooming and tagging.
    • Despite the recent strong success of agentic MLLMs, this work uncovers a critical safety failure in the tool-use paradigm: agentic tool-using MLLMs become less capable of refusing harmful requests.
    • Our experiments confirm that, across three popular safety benchmarks, all the top open- and closed-weight MLLMs we test exhibit significantly lower safety in tool-using settings than in non-tool settings, with a relative refusal failure rate increase of up to 68.7%.
  • EN 要点:
    • arXiv:2610.03938v1 Announce Type: new
    • Abstract: Agentic multimodal large language models (MLLMs) have recently pushed the frontier of visual reasoning by calling tools such as zooming and tagging
    • Despite the recent strong success of agentic MLLMs, this work uncovers a critical safety failure in the tool-use paradigm: agentic tool-using MLLMs become less…
    • Our experiments confirm that, across three popular safety benchmarks, all the top open- and closed-weight MLLMs we test exhibit significantly lower safety in to…

• Retrieval-Augmented Large Language Model Decision-Making for Autonomous Driving Guided by Chinese Philosophical Wisdom

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03948v1 Announce Type: new.
    • Abstract: Autonomous driving decision systems must balance safety, efficiency, and social norms in complex traffic interactions.
    • Philosophical and ethical considerations have received limited attention in existing autonomous driving decision-making approaches based on numerical optimization, sequence prediction, and large language models (LLMs).
    • We propose Chinese Philosophical Wisdom-Guided Driving (CPW-Drive), a closed-loop retrieval-augmented generation (RAG) framework that incorporates value guidance derived from Chinese philosophy into autonomous driving decision-making.
  • EN 要点:
    • arXiv:2610.03948v1 Announce Type: new
    • Abstract: Autonomous driving decision systems must balance safety, efficiency, and social norms in complex traffic interactions
    • Philosophical and ethical considerations have received limited attention in existing autonomous driving decision-making approaches based on numerical optimizati…
    • We propose Chinese Philosophical Wisdom-Guided Driving (CPW-Drive), a closed-loop retrieval-augmented generation (RAG) framework that incorporates value guidanc…

• LatentQuant: Preserving the Policy-Facing Latent Contract under NVFP4 VAE Quantization

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03959v1 Announce Type: new.
    • Abstract: Recent world action models (WAMs) reuse pretrained video VAEs whose encoder latents directly condition downstream action policies.
    • Quantization must therefore preserve not only reconstruction fidelity but also the policy-facing latent contract expected by the frozen policy.
    • Direct NVFP4 leaves W4A4 quantization error uncompensated, whereas joint quantization-aware training (QAT) can recover reconstruction by moving this representation.
  • EN 要点:
    • arXiv:2610.03959v1 Announce Type: new
    • Abstract: Recent world action models (WAMs) reuse pretrained video VAEs whose encoder latents directly condition downstream action policies
    • Quantization must therefore preserve not only reconstruction fidelity but also the policy-facing latent contract expected by the frozen policy
    • Direct NVFP4 leaves W4A4 quantization error uncompensated, whereas joint quantization-aware training (QAT) can recover reconstruction by moving this representat…

• ROAR: Unifying Runs across Heterogeneous AI-Driven Research Systems

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03966v1 Announce Type: new.
    • Abstract: Each run of an AI-driven research system (ADRS) is an expensive search over a vast solution space, and dependable evaluation requires many runs, making run data both costly to produce and valuable to retain for large-scale analysis.
    • Yet this data remains fragmented: teams operate in isolation, ADRS frameworks emit results in different formats, and no shared infrastructure exists to aggregate or compare runs across problems and systems.
    • We present ROAR, a solution for systematically unifying and analyzing heterogeneous ADRS outputs.
  • EN 要点:
    • arXiv:2610.03966v1 Announce Type: new
    • Abstract: Each run of an AI-driven research system (ADRS) is an expensive search over a vast solution space, and dependable evaluation requires many runs, makin…
    • Yet this data remains fragmented: teams operate in isolation, ADRS frameworks emit results in different formats, and no shared infrastructure exists to aggregat…
    • We present ROAR, a solution for systematically unifying and analyzing heterogeneous ADRS outputs

• Teaching Agents to Code Reliably

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03984v1 Announce Type: new.
    • Abstract: Autonomous coding agents solve repository issues by reading code, running commands, editing files, and submitting patches.
    • Extra inference-time compute yields gains only when it produces a useful repair and supplies reliable evidence for choosing one.
    • Three behaviors decide both, and we argue they are teachable rather than byproducts of scale, so a policy can carry them instead of a scaffold.
  • EN 要点:
    • arXiv:2610.03984v1 Announce Type: new
    • Abstract: Autonomous coding agents solve repository issues by reading code, running commands, editing files, and submitting patches
    • Extra inference-time compute yields gains only when it produces a useful repair and supplies reliable evidence for choosing one
    • Three behaviors decide both, and we argue they are teachable rather than byproducts of scale, so a policy can carry them instead of a scaffold

• Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.03998v1 Announce Type: new.
    • Abstract: Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents.
    • Their validity as substitutes for particular respondents depends on whether they reproduce individuals’ decisions.
    • We examine what information helps synthetic respondents predict each individual’s later choices, using five conditions that add progressively richer information: no personal information, demographics, personality traits, cognitive scores, and finally the respondent’s earlier survey choices as behavioral history.
  • EN 要点:
    • arXiv:2610.03998v1 Announce Type: new
    • Abstract: Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents
    • Their validity as substitutes for particular respondents depends on whether they reproduce individuals’ decisions
    • We examine what information helps synthetic respondents predict each individual’s later choices, using five conditions that add progressively richer information…

ArXiv cs.CL (B_intro+search) 链接到标题

• Zero-Shot Visualization: Exploring Text Corpora with User-Prompted Axes

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06889v1 Announce Type: new.
    • Abstract: We study the application of large language models (LLMs) to the visual exploration of textual corpora.
    • We introduce zero-shot visualization (ZSV), a task in which users specify concepts in natural language and documents are mapped onto the corresponding concept axes for visualization.
    • Building a ZSV system of practical value is non-trivial, as it requires choices at the intersection of feature functions, efficient implementation tradeoffs, and pre/post-processing decisions affecting visualization quality.
  • EN 要点:
    • arXiv:2610.06889v1 Announce Type: new
    • Abstract: We study the application of large language models (LLMs) to the visual exploration of textual corpora
    • We introduce zero-shot visualization (ZSV), a task in which users specify concepts in natural language and documents are mapped onto the corresponding concept a…
    • Building a ZSV system of practical value is non-trivial, as it requires choices at the intersection of feature functions, efficient implementation tradeoffs, an…

• Capacity, Responsiveness and Alignment: What Makes a Latent Structure Actionable

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06897v1 Announce Type: new.
    • Abstract: Localizing latent structures in the activation space of language models (LMs) is central to understanding and controlling their behavior.
    • Yet, localized structures can differ substantially in their causal influence, raising the question of what makes a structure actionable.
    • We tackle this question by casting causal influence as a product of three factors and showing empirically that they act as interpretable, distinct constraints: capacity, measuring the sensitivity of the model’s output to movement along the structure, responsiveness, capturing how promotable the concept is given the current context, and alignment, reflecting how well the structure aligns with the context-specific representation of the concept.
  • EN 要点:
    • arXiv:2610.06897v1 Announce Type: new
    • Abstract: Localizing latent structures in the activation space of language models (LMs) is central to understanding and controlling their behavior
    • Yet, localized structures can differ substantially in their causal influence, raising the question of what makes a structure actionable
    • We tackle this question by casting causal influence as a product of three factors and showing empirically that they act as interpretable, distinct constraints:…

• Tree Navigation Without LLM Summaries: A Matched-Cost Study of Hierarchical Retrieval for Long-Document QA

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06902v1 Announce Type: new.
    • Abstract: Retrieval-augmented generation grounds language models in external context, but for long documents flat top-$k$ retrieval can cluster on a single region and miss complementary evidence.
    • RAPTOR-style summary trees address this by recursively clustering chunks and using a language model to summarize each cluster at indexing time, then ranking summary nodes alongside raw chunks at query time.
    • We show the main benefit of summary trees in long-document QA can come from navigation rather than the generated summary content.
  • EN 要点:
    • arXiv:2610.06902v1 Announce Type: new
    • Abstract: Retrieval-augmented generation grounds language models in external context, but for long documents flat top-$k$ retrieval can cluster on a single regi…
    • RAPTOR-style summary trees address this by recursively clustering chunks and using a language model to summarize each cluster at indexing time, then ranking sum…
    • We show the main benefit of summary trees in long-document QA can come from navigation rather than the generated summary content

• Component and Dimension Sparsity in Transformer Refusal Mechanisms

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06903v1 Announce Type: new.
    • Abstract: Activation steering manipulates large language model behavior by intervening on internal activations, but the mechanistic basis of these interventions remains poorly understood.
    • We decompose refusal steering into component-level interventions across four open-weight models, identifying the sparse subsets of attention and MLP components whose steering suffices to reproduce the full behavioral effect.
    • We find that refusal directions concentrate in sparse component mechanisms comprising 28–48% of upstream components, retaining 88–101% of steering effectiveness.
  • EN 要点:
    • arXiv:2610.06903v1 Announce Type: new
    • Abstract: Activation steering manipulates large language model behavior by intervening on internal activations, but the mechanistic basis of these interventions…
    • We decompose refusal steering into component-level interventions across four open-weight models, identifying the sparse subsets of attention and MLP components…
    • We find that refusal directions concentrate in sparse component mechanisms comprising 28–48% of upstream components, retaining 88–101% of steering effective…

• Stabilizing language models under continual learning via condition-anchored distillation

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06940v1 Announce Type: new.
    • Abstract: Continual adaptation of language models can change their output distribution on prompts learned earlier, while retaining every old prompt-answer pair may be undesirable or impossible.
    • We study condition-anchored generative distillation (CAGD): retain a small set of old prompts, use a frozen previous model to reconstruct completions and generation states, and match its predictive distributions while learning the next task.
    • The formulation separates three roles that ordinary replay conflates: conditions select the behavior to protect, teacher generations locate relevant states, and soft targets specify how predictions may change.
  • EN 要点:
    • arXiv:2610.06940v1 Announce Type: new
    • Abstract: Continual adaptation of language models can change their output distribution on prompts learned earlier, while retaining every old prompt-answer pair…
    • We study condition-anchored generative distillation (CAGD): retain a small set of old prompts, use a frozen previous model to reconstruct completions and genera…
    • The formulation separates three roles that ordinary replay conflates: conditions select the behavior to protect, teacher generations locate relevant states, and…

• EMODE: Dynamic Para-Semantic Experts for Emotion-Aware Speech Language Modeling

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06956v1 Announce Type: new.
    • Abstract: Large speech language models have demonstrated strong capabilities in unified cross-modal understanding and generation, yet paralinguistic cues, especially emotion, remain difficult to preserve.
    • Existing systems typically rely on entangled acoustic representations, which allow the underlying language model to depend excessively on recovered lexical content instead of grounding its behavior in acoustic-prosodic evidence.
    • We address this limitation with EMODE, an emotion-aware speech language model built around \textbf{Dynamic Para-Semantic Experts (DPSE)}.
  • EN 要点:
    • arXiv:2610.06956v1 Announce Type: new
    • Abstract: Large speech language models have demonstrated strong capabilities in unified cross-modal understanding and generation, yet paralinguistic cues, espec…
    • Existing systems typically rely on entangled acoustic representations, which allow the underlying language model to depend excessively on recovered lexical cont…
    • We address this limitation with EMODE, an emotion-aware speech language model built around \textbf{Dynamic Para-Semantic Experts (DPSE)}

• Verdicts Without Annotated Evidence: Rejection Sampling or Label-Only Post-Training for Evidence Recovery?

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06962v1 Announce Type: new.
    • Abstract: In many review workflows the verdict is the only thing retained.
    • The passages behind it are not marked, because that annotation costs far more than recording the decision.
    • We measure how much of that evidence a small language model can recover when it is post-trained on the verdicts alone, with no human evidence labels at any stage.
  • EN 要点:
    • arXiv:2610.06962v1 Announce Type: new
    • Abstract: In many review workflows the verdict is the only thing retained
    • The passages behind it are not marked, because that annotation costs far more than recording the decision
    • We measure how much of that evidence a small language model can recover when it is post-trained on the verdicts alone, with no human evidence labels at any stag…

• WavePrune: One period is often enough for RoPE

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06963v1 Announce Type: new.
    • Abstract: Rotary Position Embedding (RoPE) encodes token positions by rotating each two-dimensional channel of the query and key vectors at a channel-specific frequency, making the attention logits invariant to a common shift of positions.
    • However, this rotation is periodic, and it leads to position aliasing where relative positions separated by a full rotation period become hard to tell apart.
    • To address this, we propose WavePrune, which restricts each channel to its first rotation period.
  • EN 要点:
    • arXiv:2610.06963v1 Announce Type: new
    • Abstract: Rotary Position Embedding (RoPE) encodes token positions by rotating each two-dimensional channel of the query and key vectors at a channel-specific f…
    • However, this rotation is periodic, and it leads to position aliasing where relative positions separated by a full rotation period become hard to tell apart
    • To address this, we propose WavePrune, which restricts each channel to its first rotation period

• Calibrated Answers About Randomized Trials From a 4-Billion-Parameter Open Model: A Registered Test and a License-Clean Release

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.07019v1 Announce Type: new.
    • Abstract: Fiorillo v0.5 is an open model that answers typed questions with a probability for each answer.
    • Its main specialist reads a randomized trial’s article, cut to 6,144 tokens, and answers whether an intervention significantly increased, significantly decreased or did not significantly change an outcome against a comparator (Evidence Inference 2.0, EI).
    • It is Qwen3-4B-Base with low-rank adapters and a decision head, fine-tuned for EI only on the 1,431 of 2,657 training articles whose own license allows reuse.
  • EN 要点:
    • arXiv:2610.07019v1 Announce Type: new
    • Abstract: Fiorillo v0.5 is an open model that answers typed questions with a probability for each answer
    • Its main specialist reads a randomized trial’s article, cut to 6,144 tokens, and answers whether an intervention significantly increased, significantly decrease…
    • It is Qwen3-4B-Base with low-rank adapters and a decision head, fine-tuned for EI only on the 1,431 of 2,657 training articles whose own license allows reuse

• Investigating Model Compression for Neural Machine Translation in the Biomedical Domain

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.07032v1 Announce Type: new.
    • Abstract: Large-scale pretrained transformer models have achieved state-of-the-art performance across diverse machine translation tasks, including multilingual settings.
    • Knowledge distillation has emerged as a sustainable approach for model compression, transferring knowledge from large teacher models to smaller, more efficient student models.
    • Similarly, quantization, which reduces the numerical precision of model weights and activations (e.g., from 32-bit to 8-bit representations) is widely used to accelerate inference, enabling models to run several times faster during deployment.
  • EN 要点:
    • arXiv:2610.07032v1 Announce Type: new
    • Abstract: Large-scale pretrained transformer models have achieved state-of-the-art performance across diverse machine translation tasks, including multilingual…
    • Knowledge distillation has emerged as a sustainable approach for model compression, transferring knowledge from large teacher models to smaller, more efficient…
    • Similarly, quantization, which reduces the numerical precision of model weights and activations (e.g., from 32-bit to 8-bit representations) is widely used to a…

ArXiv cs.LG (B_intro+search) 链接到标题

• TEMPEST: Temporal Embeddings for Scalable Driver Identification via Angular Margin Learning

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06855v1 Announce Type: new.
    • Abstract: Scalable driver identification requires embedding models that maintain discriminative performance as fleet size grows, yet existing triplet-loss formulations degrade rapidly with driver pool size and overfit to session-specific patterns under rigorous temporal evaluation.
    • We introduce TEMPEST, a Temporal Convolutional Network embedding model trained with an additive angular margin (ArcFace) loss that enforces global class-level separation in a normalized angular space.
    • TEMPEST maps 60-second multimodal driving windows to compact 96-dimensional embeddings, supporting truly dynamic enrollment without any retraining or classifier refitting.
  • EN 要点:
    • arXiv:2610.06855v1 Announce Type: new
    • Abstract: Scalable driver identification requires embedding models that maintain discriminative performance as fleet size grows, yet existing triplet-loss formu…
    • We introduce TEMPEST, a Temporal Convolutional Network embedding model trained with an additive angular margin (ArcFace) loss that enforces global class-level s…
    • TEMPEST maps 60-second multimodal driving windows to compact 96-dimensional embeddings, supporting truly dynamic enrollment without any retraining or classifier…

• When Does External Guidance Help LLM Reasoning? A Bias-Variance Theory of Guidance-Augmented GRPO

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06861v1 Announce Type: new.
    • Abstract: Reinforcement learning with verifiable rewards (RLVR) has become the dominant paradigm for eliciting multi-step reasoning in large language models, and a recent wave of methods (LUFFY, ExPO, PAPO, TAPO) further augments RL with \emph{external guidance} - expert traces, self-explanations, or retrieved thought patterns.
    • Although each method reports empirical gains, none provides convergence rates, bias bounds, or an optimal weighting rule for the guidance signal.
    • We close this gap with \emph{Guidance-Augmented GRPO} (GA-GRPO), a unified theoretical framework that casts external guidance as a stochastic guidance operator G re-writing the question distribution, and analyses the resulting policy-gradient estimator as a biased on-policy estimator whose bias is bounded by the total-variation guidance divergence delta_G between the guidance-augmented sampling distribution and the policy’s own distribution.
  • EN 要点:
    • arXiv:2610.06861v1 Announce Type: new
    • Abstract: Reinforcement learning with verifiable rewards (RLVR) has become the dominant paradigm for eliciting multi-step reasoning in large language models, an…
    • Although each method reports empirical gains, none provides convergence rates, bias bounds, or an optimal weighting rule for the guidance signal
    • We close this gap with \emph{Guidance-Augmented GRPO} (GA-GRPO), a unified theoretical framework that casts external guidance as a stochastic guidance operator…

• Neutrosophic Ensemble Classification for Uncertainty-Aware Bearing Fault Detection: Evidence from Laboratory and Variable-Speed Industrial Benchmarks

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06880v1 Announce Type: new.
    • Abstract: Machine learning classifiers for bearing fault detection produce scalar confidence scores that conflate confident errors with genuinely ambiguous predictions, and the conventional truth/falsity pair (F = 1 - T) is algebraically redundant by construction.
    • We operationalize a refined neutrosophic decomposition of a Random Forest + XGBoost + Logistic Regression ensemble into four indicators – T-hat (top-class evidence), F-hat (best-competitor evidence), predictive entropy I1-hat, and decision disagreement I2-hat – evaluated on two bearing benchmarks (CWRU and JNU, 600-1000 rpm) under a leave-one-condition-out protocol.
    • On CWRU, after correcting a file-to-class mapping error, the ensemble reaches 100.00 percent accuracy on three of four held-out loads (92.27 percent on the fourth), leaving too few errors for uncertainty analysis.
  • EN 要点:
    • arXiv:2610.06880v1 Announce Type: new
    • Abstract: Machine learning classifiers for bearing fault detection produce scalar confidence scores that conflate confident errors with genuinely ambiguous pred…
    • We operationalize a refined neutrosophic decomposition of a Random Forest + XGBoost + Logistic Regression ensemble into four indicators – T-hat (top-class evid…
    • On CWRU, after correcting a file-to-class mapping error, the ensemble reaches 100.00 percent accuracy on three of four held-out loads (92.27 percent on the four…

• Comparative review of hybrid forecasting models for short-term prediction of building thermal load

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06881v1 Announce Type: new.
    • Abstract: In this paper, a comparative review of different hybrid models for short-term forecasting of building thermal demand is carried out.
    • Particularly, the assessment tackles the comparison of data-driven models enhanced with other state-of-the-art techniques.
    • At the first step, the existing techniques reported in the literature are analysed.
  • EN 要点:
    • arXiv:2610.06881v1 Announce Type: new
    • Abstract: In this paper, a comparative review of different hybrid models for short-term forecasting of building thermal demand is carried out
    • Particularly, the assessment tackles the comparison of data-driven models enhanced with other state-of-the-art techniques
    • At the first step, the existing techniques reported in the literature are analysed

• Learning When to Refine: Long-Horizon Reinforcement Learning for Budgeted Neural-Operator PDE Solvers

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06883v1 Announce Type: new.
    • Abstract: Neural operators provide fast surrogates for time-dependent PDEs, but autoregressive deployment creates a refinement-allocation problem: prediction errors vary over space and time, while only a finite number of local corrections can be committed along a trajectory.
    • We formulate this as budgeted adaptive neural-operator solving.
    • A global Fourier neural operator advances the full field, a local operator proposes patch-wise residual corrections, and a set-aware selector chooses where to refine.
  • EN 要点:
    • arXiv:2610.06883v1 Announce Type: new
    • Abstract: Neural operators provide fast surrogates for time-dependent PDEs, but autoregressive deployment creates a refinement-allocation problem: prediction er…
    • We formulate this as budgeted adaptive neural-operator solving
    • A global Fourier neural operator advances the full field, a local operator proposes patch-wise residual corrections, and a set-aware selector chooses where to r…

• Event-Driven ML Pipeline Orchestration for Manufacturing: An AWS Industry Experience

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06890v1 Announce Type: new.
    • Abstract: We present an industry experience report on three years of operating an event-driven cloud infrastructure for continuous machine learning training in automotive manufacturing.
    • Our system orchestrates GPU-accelerated training of product-specialized model pairs, a physics prediction model and a reinforcement-learning control policy, across multiple plants, coordinating long-running GPU workloads triggered by manufacturing events.
    • The architecture combines Amazon ECS with EC2 GPU capacity providers, SQS-based messaging with dead-letter queues, and an admission-controlled Lambda dispatcher that enforces cluster concurrency limits.
  • EN 要点:
    • arXiv:2610.06890v1 Announce Type: new
    • Abstract: We present an industry experience report on three years of operating an event-driven cloud infrastructure for continuous machine learning training in…
    • Our system orchestrates GPU-accelerated training of product-specialized model pairs, a physics prediction model and a reinforcement-learning control policy, acr…
    • The architecture combines Amazon ECS with EC2 GPU capacity providers, SQS-based messaging with dead-letter queues, and an admission-controlled Lambda dispatcher…

• Learning from Unreliable Trajectories: Adversarially-Robust Federated Q-Learning

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06918v1 Announce Type: new.
    • Abstract: We study federated reinforcement learning in which multiple agents interact with a common Markov decision process and communicate through a central server to collaboratively learn the optimal state-action value function.
    • Our goal is to understand whether the sample-efficiency benefits of collaboration can be retained when a fraction of the agents behave adversarially and transmit arbitrarily corrupted information.
    • To address this problem, we introduce Robust Async-Fed-Q, an epoch-based federated learning algorithm that combines variance-reduced estimation of the Bellman optimality operator at the agents with robust aggregation at the server.
  • EN 要点:
    • arXiv:2610.06918v1 Announce Type: new
    • Abstract: We study federated reinforcement learning in which multiple agents interact with a common Markov decision process and communicate through a central se…
    • Our goal is to understand whether the sample-efficiency benefits of collaboration can be retained when a fraction of the agents behave adversarially and transmi…
    • To address this problem, we introduce Robust Async-Fed-Q, an epoch-based federated learning algorithm that combines variance-reduced estimation of the Bellman o…

• AttSVD:Prompt-Adaptive Low-Rank KV Cache Compression via Attention-Guided SVD

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06927v1 Announce Type: new.
    • Abstract: The key-value (KV) cache of autoregressive transformers grows linearly with context length and dominates memory at long context.
    • Most training-free remedies evict low-importance tokens, an irreversible choice along the sequence axis.
    • We instead keep every token and store it more cheaply along the “feature” axis.
  • EN 要点:
    • arXiv:2610.06927v1 Announce Type: new
    • Abstract: The key-value (KV) cache of autoregressive transformers grows linearly with context length and dominates memory at long context
    • Most training-free remedies evict low-importance tokens, an irreversible choice along the sequence axis
    • We instead keep every token and store it more cheaply along the “feature” axis

• Near-Optimal Sample Complexity for Recursive Entropic Risk Reinforcement Learning with a Generative Model

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06931v1 Announce Type: new.
    • Abstract: In this paper, we study the sample complexities of value and policy learning in finite discounted Markov decision processes (MDPs) under recursive entropic risk preferences with risk parameter (\beta\neq 0), assuming access to a generative model of the MDP.
    • We provide a refined analysis of model-based risk-sensitive Q-value iteration (MB-RS-QVI), a plug-in model-based method introduced in prior work, and derive ((\varepsilon,\delta))-PAC guarantees for both learning the optimal (Q)-value function and an (\varepsilon)-optimal policy.
    • Our bounds improve the exponential dependence on the effective horizon (1/(1-\gamma)) compared with the best existing guarantees for this setting.
  • EN 要点:
    • arXiv:2610.06931v1 Announce Type: new
    • Abstract: In this paper, we study the sample complexities of value and policy learning in finite discounted Markov decision processes (MDPs) under recursive ent…
    • We provide a refined analysis of model-based risk-sensitive Q-value iteration (MB-RS-QVI), a plug-in model-based method introduced in prior work, and derive ((…
    • Our bounds improve the exponential dependence on the effective horizon (1/(1-\gamma)) compared with the best existing guarantees for this setting

• QiYao-I: A Manifold Based Foundation Model for Irregular Multivariate Time Series Forecasting

  • 发布时间:2026-10-07 12:00 北京时间
  • 摘要:【待翻译】- arXiv:2610.06936v1 Announce Type: new.
    • Abstract: Irregular multivariate time series forecasting is a challenging yet important problem in real-world applications, where observations are often irregularly sampled and asynchronously recorded across variables.
    • Existing time series foundation models are mostly built on regularly sampled sequences, making them difficult to generalize to irregular time intervals and asynchronous cross-variable dependencies.
    • To address these challenges, we propose QiYao-I, a manifold based foundation model for irregular multivariate time series forecasting.
  • EN 要点:
    • arXiv:2610.06936v1 Announce Type: new
    • Abstract: Irregular multivariate time series forecasting is a challenging yet important problem in real-world applications, where observations are often irregul…
    • Existing time series foundation models are mostly built on regularly sampled sequences, making them difficult to generalize to irregular time intervals and asyn…
    • To address these challenges, we propose QiYao-I, a manifold based foundation model for irregular multivariate time series forecasting