System translated (Gemini)

🤖 AI 速览

Today’s Core: Agents are entering real consumer workflows—ChatGPT for Teens has launched College Planner, flashcards, and quizzes, while Radisson hotel booking now integrates with ChatGPT. Stratechery analyzes the Apple/LG smart home and Amazon’s agent standard. In life sciences, …
📋 文章元数据
发布时间
2026-10-08
类型
ai-daily
字数
6604
阅读时长
32 min

2026-10-08 AI Daily | ChatGPT Expands to Teen Planning and Hotel Bookings, DeepMind’s AlphaGenome Atlas Leads Calibratable Life Sciences Link to heading

Today’s Core: Agents are entering real-world consumer workflows—ChatGPT for Teens launches College Planner, Flashcards, and Quizzes, while Radisson hotel bookings are now integrated with ChatGPT. Stratechery analyzes the Apple/LG smart home ecosystem and Amazon’s agent standards. In life sciences, DeepMind’s AlphaGenome Atlas and calibratable models for randomized trials highlight a trend toward traceability and calibration. arXiv sees a dense wave of updates on interpretability.

📖 Deep Dive on This Issue’s Watch List Link to heading

Three key themes are worth a close read today. First, agents are moving from demos to real workflows: OpenAI has launched a College Planner, Flashcards, and Quizzes for teenagers; Radisson, in partnership with Accenture, has integrated hotel discovery and booking into ChatGPT; and Stratechery’s analysis of the Apple/LG smart home and Amazon’s agent standards is worth the attention of product and ecosystem teams. Second, interpretability is seeing a wave of systematic updates: a concentration of arXiv papers on the sparsity of Transformer refusal mechanisms, the manipulability of latent structures, RoPE pruning, and stabilization of continual learning has emerged, which is relevant for alignment and infrastructure teams to track. Third, in life sciences and specialized evidence: the widely discussed DeepMind update points to the AlphaGenome Atlas. Additionally, a 4B open model providing calibrated probabilities for randomized trials and advancements in biomedical translation compression indicate that “traceability and calibration” are becoming new benchmarks for domain-specific models.

🌐 AI Hot Topics on X Link to heading

Topic 1: Grok Bot Routes Tasks to Top AI Models for Best Results Link to heading

  • Category: AI · News
  • Overview: Trending: 2 days ago, Related posts: 64,000
  • What it is: The Grok bot has started automatically routing user tasks to multiple top AI models to return the best results.
  • Why it’s important: This indicates a shift in AI assistants from competing on single-model capabilities to focusing on multi-model scheduling and optimization, which could improve response quality and reshape the model ecosystem.
  • Discussion summary: Discussions on X focus on whether the routing mechanism can truly select the optimal model, response speed and cost, and the practical gap compared to products like ChatGPT and Claude.

Topic 2: Gemini 4 Argon Appears in Google Services Ahead of Launch Link to heading

  • Category: AI · News
  • Overview: Trending: 1 day ago, Related posts: 3,500
  • What it is: Google’s unreleased Gemini 4 Argon model has appeared in Google services, drawing attention to its upcoming launch.
  • Why it’s important: This is seen as a signal ahead of the release of Google’s next-generation flagship model, which could impact the competitive landscape of AI models and industry benchmark evaluations.
  • Discussion summary: There is disagreement on X regarding its benchmark performance: the model won 13 out of 19 tests, but the margins in the tests it lost were larger. The debate centers on whether the number of wins reflects a true lead in capability.

Topic 3: OpenAI Releases 722 AI-Generated Math Breakthroughs on GitHub Link to heading

  • Category: AI · News
  • Overview: Trending: 1 day ago, Related posts: 88,000
  • What it is: OpenAI has released 722 AI-generated mathematical breakthroughs on GitHub.
  • Why it’s important: This demonstrates the advancing capabilities of AI in mathematical reasoning and scientific discovery, potentially accelerating research progress and changing scientific methods.
  • Discussion summary: Discussions on X focus on whether these results have been rigorously verified, whether they can be considered true mathematical breakthroughs, and the impact of AI-generated content on academic research and researchers.

Topic 4: Germany Pushes for EU Approval of Tesla’s Full Self-Driving Supervised Link to heading

  • Category: AI · News
  • Overview: Trending: 17 hours ago, Related posts: 20,000
  • What it is: Germany is pushing for EU approval of Tesla’s Full Self-Driving (FSD) Supervised system.
  • Why it’s important: If approved, Tesla’s driver-assistance technology would enter the core EU market for the first time, potentially accelerating the evolution of autonomous driving regulations and the competitive landscape in Europe.
  • Discussion summary: The focus on X includes the influence of Germany’s regulatory stance on other EU member states, the compatibility of Tesla’s FSD with existing European safety standards, and whether this move will put pressure on local automakers.

Topic 5: Ben Affleck Reveals Python Coding and Neural Network Skills from VFX Work Link to heading

  • Category: AI · Entertainment
  • Overview:Trending since:1 day ago,Related posts:16000
  • What happened:Ben Affleck revealed that he learned Python programming and neural network skills while working in visual effects.
  • Why it matters:An A-list Hollywood actor publicly discussing programming and neural networks could increase public attention on the application of AI in film special effects and content creation.
  • Discussion summary:Discussions on X centered on surprise and skepticism about the celebrity’s technical skills, the significance of stars getting involved in AI and film production, and whether programming and neural networks are becoming essential tools in the creative industry.

AI Public Opinion Summary on X Today Link to heading

The main theme in today’s public opinion is that AI is transitioning from a competition based on single-model capabilities to a new phase characterized by multi-model orchestration, jockeying for position with next-generation models, and accelerated penetration into vertical industries. This trend is exemplified by Grok’s multi-model routing, the early leak of Google’s Gemini 4 Argon, and OpenAI’s mathematical achievements. The consensus is that multi-model collaboration and more advanced reasoning capabilities will improve answer quality and boost efficiency in scientific research and creative industries. However, significant disagreements are also apparent: there are debates on whether Grok can truly select the optimal model while balancing cost and speed, whether Gemini’s lead in multiple areas constitutes a true advantage given its significant lag in others, whether OpenAI’s 722 mathematical achievements can withstand rigorous verification, and whether Tesla’s FSD meets European safety expectations. Potential risks are concentrated in several areas: unverified AI research findings could undermine academic trust; relaxed autonomous driving regulations could create safety issues and pressure local industries; and the costs of model orchestration, along with the halo effect of celebrity tech involvement, might exacerbate ecosystem dependencies or public misunderstanding.

💡 Influencer Insights Link to heading

Influencer insights are unavailable today. We recommend reading the in-depth content from the Watch List.

📚 Appendix: Today’s Watch List Update Source List Link to heading

Timeframe: Last 3 days; 22 sources covered; 35 updates in total

Stratechery by Ben Thompson (A_full) Link to heading

  • Apple and LG, The House For Everyone Else, Agent Standards and Amazon
    • Published: 2026-10-07 18:01 Beijing Time
    • Abstract: [NEEDS TRANSLATION] - 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 Key Points:
      • 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) Link to heading

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

    • Published: 2026-10-07 20:00 Beijing Time
    • Abstract: [NEEDS TRANSLATION] - 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.

    • Key Points:
      • 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

    • Published: 2026-10-07 15:00 Beijing Time
    • Summary: 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.
    • Key Points:
      • 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

    • Published: 2026-10-07 08:00 Beijing Time
    • Summary: 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.
    • Key Points:
      • 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) Link to heading

  • DeepMind’s New AI Just Cracked The Code Of Life
    • Published: 2026-10-07 21:29 Beijing Time
    • Summary: ❤️ 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 Key Points:
    • ❤️ 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) Link to heading

  • Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Key Points:

    • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Pending Translation] - 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 Key Points:
      • 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

    • Release Time: 2026-10-07 12:00 Beijing Time
    • Summary: [To be translated] - 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 Highlights:
      • 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

    • Release Time: 2026-10-07 12:00 Beijing Time
    • Summary: [To be translated] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Awaiting Translation] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Highlights:
      • 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

    • Published: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Highlights:
      • 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

    • Published: 2026-10-07 12:00 Beijing Time

  • Abstract: [Translation Pending] - 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 Key Points:
      • 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) Link to heading

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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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

    • Published: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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

    • Published: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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…

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

    • Release Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Pending Translation] - 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 Key Points:
      • 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?

    • Release Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Pending Translation] - 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 Key points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Key points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Highlights:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Awaiting Translation] - 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 Highlights:
      • 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) Link to heading

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

    • Release Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [TO BE TRANSLATED] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Highlights:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [Translation Pending] - 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract:【To be translated】- 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract:【To be translated】- 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 Key Points:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Highlights:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Highlights:
      • 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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: [To be translated] - 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 Key points:
      • 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

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

    • Publication Time: 2026-10-07 12:00 Beijing Time
    • Abstract: 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.

  • Key Points:

    • 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