System translated (Gemini)

🤖 AI 速览

Today’s main theme focuses on the governance of AI programming tools: ZCode is accused of uploading entire workspaces, sparking discussions on code privacy and informed consent; while Claude Code’s support for AGENTS.md shows that team-level AI collaboration is moving towards …
📋 文章元数据
发布时间
2026-09-19
类型
ai-daily
字数
6290
阅读时长
30 min

2026-09-19 AI Daily | When AI Starts Reading the Entire Workspace: The Next Level for Programming Assistants is Transparency and Standardization Link to heading

Today’s main focus is on AI programming tool governance: ZCode is accused of uploading entire workspaces, sparking discussions around code privacy and informed consent; Claude Code’s support for AGENTS.md, meanwhile, indicates that team-level AI collaboration is moving towards standardization. At the same time, local-first agents and model trustworthiness research continue to heat up, with industry attention shifting from “can it be used” to “can it be trusted and managed.”

📖 In-depth Guide to This Issue’s Watch List Link to heading

Today’s Watch List can be viewed along three lines.

First is the industry and capital narrative: Stratechery’s weekly report and “The State of Startups in 2026” should be read together. The former looks at major platforms and the business landscape, while the latter observes new opportunities for startups in local AI assistants, open source, and distribution.

The second line is AI trustworthiness and user experience. Papers such as “App Store Generative AI Review Analysis,” “FakeSpotter,” and “Implicit Conflict Detection” all address the same question: models must not only be “usable” but also trusted, understood, and reduce misguidance.

The third line is the deepening of model internal mechanisms and evaluation. Subconscious prompting, style axis discovery, self-reflective reasoning, interdisciplinary scientific reasoning, and classical poetry evaluation show that the research focus is shifting from simply chasing scores to explaining how models express, reason, and make mistakes.

🌐 X Platform AI Hot News Link to heading

Topic 1: ZCode AI App Uploaded Full Workspaces to Alibaba Cloud Without Clear Disclosure Link to heading

  • Category: AI · News
  • Overview: Trending 15 hours ago, 7200 related posts
  • What happened: There are reports that the ZCode AI application uploaded users’ complete workspaces to Alibaba Cloud without clear disclosure.
  • Why it’s important: This incident involves the scope of data collection by AI programming tools, code privacy, cloud storage, and user informed consent, potentially affecting developers’ trust in the security and compliance of AI tools.
  • Discussion overview: Discussions on X mainly focus on whether the application actually uploaded complete workspaces, whether the privacy policy is consistent with actual behavior, whether users were fully informed, and the security and compliance risks of storing data on Alibaba Cloud; some also believe it’s necessary to await official responses and technical evidence to avoid drawing conclusions based solely on headlines.

Topic 2: Claude Code Now Supports AGENTS.md Files for Better Team Workflows Link to heading

  • Category: AI · News
  • Overview: Trending 4 hours ago, 6300 related posts
  • What happened: Anthropic’s Claude Code now supports AGENTS.md files, making it easier for teams to uniformly configure project descriptions, specifications, and workflows for AI programming agents.
  • Why it’s important: This indicates that AI programming tools are moving from individual assistance to team collaboration scenarios, emphasizing reusable context, standardized instructions, and integration into engineering processes, which is crucial for improving code generation quality and organizational adoption.
  • Discussion overview: Discussions on X primarily center on whether AGENTS.md will become a common convention for AI programming agents, its value compared to existing READMEs or prompt configurations, and the actual benefits and potential complexities for teams in terms of permissions, code quality, and process governance.

Today’s AI Public Opinion Summary on X Link to heading

Today’s main public opinion focuses on the necessary concurrent improvement of data governance and engineering standards as AI programming tools transition from “personal efficiency assistants” to “team-level infrastructure.” On one hand, ZCode AI is accused of uploading entire workspaces without sufficient disclosure, raising widespread concerns about code privacy, user informed consent, cloud storage, and compliance; however, whether this behavior actually occurred still requires official responses and technical evidence. On the other hand, Claude Code’s support for AGENTS.md is seen as an important attempt to promote unified AI programming standards, context reuse, and integration into engineering processes for teams. The overall consensus is that AI programming tools require more transparent data processing mechanisms and more standardized collaboration configurations. Disagreements lie in whether specific products have crossed privacy boundaries and whether new configuration files can genuinely deliver stable quality and governance benefits. Potential risks include privacy incidents eroding developer trust, team instruction configurations causing permission or code quality issues, and increased governance complexity due to a lack of unified industry standards.

💡 Influencer Insights Link to heading

Today’s influencer insights are currently unavailable. Recommended reading: Watch List in-depth content.

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

Time window: Last 3 days; covering 22 sources; 34 updates in total

Y Combinator Podcast (B_intro+search) Link to heading

  • The State of Startups in 2026
    • Published: 2026-09-18 22:59 Beijing Time
  • Abstract: [To be translated]- You’ve probably already heard all about OpenClaw (formerly Clawdbot/Moltbot).
    • The viral sensation is an open-source AI assistant that runs on your own device, connects with messaging apps you already use, and goes beyond chat to actually execute tasks like managing your email, calendars, files, workflows, and more.
    • Now meet the man behind it.
    • YC’s Raphael Schaad sat down with Peter Steinberger, the creator of OpenClaw, to discuss the “aha” moment behind the viral personal AI agent, why local-first agents could replace many of today’s apps, and how personal agents will reshape the future of software.
    • EN Key Points:
      • YC works with thousands of founders every year, which gives us an early look at how startups are changing
      • Right now, the shift is striking: startups are moving from bits to atoms, nearly one in five YC companies has a solo founder, and companies are reaching meaning…
      • They discuss how AI is making it possible for smaller teams to take on more ambitious problems, why experienced founders are having a resurgence, and why knowin…

All-In Podcast (A_full) Link to heading

  • Jared Isaacman: A New Era for NASA and American Space Exploration

    • Publication Time: 2026-09-18 23:59 Beijing Time
    • Abstract: [To be translated]- IREN is a vertically integrated AI Cloud platform, delivering data centers, compute and software for AI training and inference.
      • Oracle connects the data, applications, and infrastructure that turn AI into business outcomes—with the flexibility, choice, and control to optimize as AI evolves.
      • EY helps tech innovators scale from startup to exit to megacap.
      • We’re focused on giving every person the tools to reach their full potential and making sure the benefits of technology are distributed to all.
      • Keel Infrastructure owns the power, land, and connectivity that HPC and AI run on - backed by secured energy assets and established grid interconnections across North America.
    • EN Key Points:

  • (0:00) NASA Administrator Jared Isaacman lays out the state and future of NASA

    • (17:59) Moon base, creating new space industries, NASA’s next chapter
    • (23:29) How to make Mars realistic
    • (25:30) Recruiting, moonshots, the “Nuclear NASA,” new vehicles and the importance of humans in space

  • Nick Shirley: Exposing Government Fraud, Suing California & Taking on the Media

    • Publication Time: 2026-09-18 08:30 Beijing Time
    • Abstract: [To be translated] - IREN is a vertically integrated AI Cloud platform, delivering data centers, compute and software for AI training and inference.
      • Oracle connects the data, applications, and infrastructure that turn AI into business outcomes—with the flexibility, choice, and control to optimize as AI evolves.
      • EY helps tech innovators scale from startup to exit to megacap.
      • We’re focused on giving every person the tools to reach their full potential and making sure the benefits of technology are distributed to all.
      • Keel Infrastructure owns the power, land, and connectivity that HPC and AI run on - backed by secured energy assets and established grid interconnections across North America.
    • EN Highlights:
      • (0:00) Nick Shirley debuts his new video: “Inside California’s High Speed Rail to Nowhere”
      • (11:42) Nick joins The Besties to discuss California’s “Stop Nick Shirley Act” and recent fraud arrests
      • (17:03) How he got into investigative journalism and finding fraud
      • (23:32) Selective journalism and why Nick is a major threat to mainstream outlets

Stratechery by Ben Thompson (A_full) Link to heading

  • 2026.38: Doomforce
    • Publication Time: 2026-09-19 01:00 Beijing Time
    • Abstract: [To be translated] - (Photo by Benjamin Fanjoy/Getty Images).
      • Welcome back to This Week in Stratechery!
      • As a reminder, each week, every Friday, we’re sending out this overview of content in the Stratechery bundle; highlighted links are free for everyone.
      • Additionally, you have complete control over what we send to you.
  • On that note, here were a few of our favorites this week.
    • EN Key Points:
      • (Photo by Benjamin Fanjoy/Getty Images)
      • Welcome back to This Week in Stratechery
      • As a reminder, each week, every Friday, we’re sending out this overview of content in the Stratechery bundle; highlighted links are free for everyone
      • Additionally, you have complete control over what we send to you

ArXiv cs.AI (B_intro+search) Link to heading

  • Regularized Emphatic Temporal-Difference Learning: Stability under Constant Stepsizes

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19170v1 Announce Type: new.
      • Abstract: Emphatic temporal-difference learning (ETD) stabilizes the expected off-policy TD update and changes its projection geometry, but neither property determines constant-stepsize sampled dynamics.
      • We construct an ergodic two-state counterexample in which the ETD mean map contracts while the sampled product has a positive top Lyapunov exponent.
      • Regenerative-cycle analysis separates this sign from the infinite variance of the follow-on trace.
    • EN Key Points:
      • arXiv:2609.19170v1 Announce Type: new
      • Abstract: Emphatic temporal-difference learning (ETD) stabilizes the expected off-policy TD update and changes its projection geometry, but neither property det…
      • We construct an ergodic two-state counterexample in which the ETD mean map contracts while the sampled product has a positive top Lyapunov exponent
      • Regenerative-cycle analysis separates this sign from the infinite variance of the follow-on trace

  • BioPhys-Bridge: A Benchmark for Interdisciplinary Scientific Reasoning in Physics-Grounded Biological Research

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19180v1 Announce Type: new.
      • Abstract: Language models face unique challenges in analyzing interdisciplinary scientific research literature.

  • In biophysics research, faithful answers require grounding observed data in source evidence, interpreting it through a quantitative physics model, and linking it to a biological mechanism.

    • To address this challenge, we introduce BioPhys-Bridge, a novel benchmark dataset for evidence-grounded scientific reasoning over biophysical literature.
    • EN Key Points:
      • arXiv:2609.19180v1 Announce Type: new
      • Abstract: Language models face unique challenges in analyzing interdisciplinary scientific research literature
      • In biophysics research, faithful answers require grounding observed data in source evidence, interpreting it through a quantitative physics model, and linking i…
      • To address this challenge, we introduce BioPhys-Bridge, a novel benchmark dataset for evidence-grounded scientific reasoning over biophysical literature

  • What Do We Expect from LLMs? Mapping the Design of LLM Benchmarks

    • Published: 2026-09-18 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.19182v1 Announce Type: new.
      • Abstract: Benchmarks are central to how progress in large language models (LLMs) is assessed and communicated.
      • Yet model rankings alone reveal little about how evaluation requirements themselves are changing.
      • The expanding variety of benchmarks offers another perspective: what researchers expect LLMs to do, and what they count as successful performance.
    • EN Key Points:
      • arXiv:2609.19182v1 Announce Type: new
      • Abstract: Benchmarks are central to how progress in large language models (LLMs) is assessed and communicated
      • Yet model rankings alone reveal little about how evaluation requirements themselves are changing
      • The expanding variety of benchmarks offers another perspective: what researchers expect LLMs to do, and what they count as successful performance

  • Position: It is Time to Virtualize Foundation Models with a Self-evolving Operating System Layer

    • Published: 2026-09-18 12:00 Beijing Time

  • Summary: - arXiv:2609.19203v1 Announce Type: new.

    • Abstract: AI applications have shifted from single, monolithic foundation models (FM) to compound agentic systems.
    • Yet today’s stacks remain fragmented: even as protocols (e.g., MCP, A2A) ease tool/agent connectivity, each framework embeds an implicit runtime for state, memory, budgets, and guardrails, making behavior non-portable and governance brittle.
    • It mirrors computing before operating systems, when every program re-implemented basic services.
    • EN Key Points:
      • arXiv:2609.19203v1 Announce Type: new
      • Abstract: AI applications have shifted from single, monolithic foundation models (FM) to compound agentic systems
      • Yet today’s stacks remain fragmented: even as protocols (e.g., MCP, A2A) ease tool/agent connectivity, each framework embeds an implicit runtime for state, memo…
      • It mirrors computing before operating systems, when every program re-implemented basic services

  • What Do Current Systematic Generalization Tasks Miss? A Reasoning-Centered Analysis

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Summary: - arXiv:2609.19212v1 Announce Type: new.
      • Abstract: Systematic generalization, the ability to solve novel problems by recombining known atomic elements, is central to human intelligence but difficult to study rigorously under controlled settings.
      • Existing studies therefore rely on simplifications such as approximately linear action composition, productivity-based tests, and action-explicit goals, which make systematic generalization easier to study but omit some essential aspects of this capability.
      • To characterize what these simplifications miss, we adopt a reasoning-centered lens and introduce TranSGrid, a testbed that brings deductive, inductive, and abductive reasoning together within a unified task.
    • EN Key Points:
      • arXiv:2609.19212v1 Announce Type: new

  • Abstract: Systematic generalization, the ability to solve novel problems by recombining known atomic elements, is central to human intelligence but difficult to…

  • Existing studies therefore rely on simplifications such as approximately linear action composition, productivity-based tests, and action-explicit goals, which m…

  • To characterize what these simplifications miss, we adopt a reasoning-centered lens and introduce TranSGrid, a testbed that brings deductive, inductive, and abd…

  • Characterizing Web Search by Conversational LLM Agents: From Search Decisions and Strategies to Results and Responses

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: arXiv:2609.19244v1 Announce Type: new.
      • Abstract: Conversational LLM agents increasingly rely on Web search, yet the end-to-end lifecycle of agentic search remains poorly understood.
      • We present the first study of Web search across four major conversational platforms (ChatGPT, Claude, Grok, and DeepSeek), combining real-world user interactions (invivo) with controlled experiments using the same platform’s models by their APIs (invitro).
      • We investigate the quality of agentic decisions to invoke Web search, their strategies to formulate queries, the potential domain preferences in the search results they receive, and the choices they make when transforming search results into grounded responses.
    • Key Points:
      • arXiv:2609.19244v1 Announce Type: new
      • Abstract: Conversational LLM agents increasingly rely on Web search, yet the end-to-end lifecycle of agentic search remains poorly understood
      • We present the first study of Web search across four major conversational platforms (ChatGPT, Claude, Grok, and DeepSeek), combining real-world user interaction…
      • We investigate the quality of agentic decisions to invoke Web search, their strategies to formulate queries, the potential domain preferences in the search resu…

  • Do AI Agents Understand Computer Architecture?

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Summary: - arXiv:2609.19387v1 Announce Type: new.
      • Abstract: Agents are increasingly asked to design hardware, and increasingly reported to succeed.
      • Such reports establish that a design improved; they cannot establish why.
      • An agent that improves an accelerator may be reasoning about the machine, or may be searching competently over knobs whose meaning it never recovers – and only the first transfers to the next architecture.
    • Key English Points:
      • arXiv:2609.19387v1 Announce Type: new
      • Abstract: Agents are increasingly asked to design hardware, and increasingly reported to succeed
      • Such reports establish that a design improved; they cannot establish why
      • An agent that improves an accelerator may be reasoning about the machine, or may be searching competently over knobs whose meaning it never recovers – and only…

  • MAGS: Multi-agent Auto-formalization Guarantees Safety for Agentic Outputs

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Summary: - arXiv:2609.19391v1 Announce Type: new.
      • Abstract: LLM coding agents now generate complex programs at a scale that makes thorough human review increasingly difficult, raising the risk of safety and security failures.
      • Common approaches, including fuzz testing, static analysis, and LLM-as-a-Verifier, can detect many failures but struggle to cover all possible edge cases.
      • Formal verification addresses this by providing machine-checkable guarantees over specified properties, but traditionally demands substantial manual specification and proof engineering.
    • Key English Points:
      • arXiv:2609.19391v1 Announce Type: new
      • Abstract: LLM coding agents now generate complex programs at a scale that makes thorough human review increasingly difficult, raising the risk of safety and sec…

  • Common approaches, including fuzz testing, static analysis, and LLM-as-a-Verifier, can detect many failures but struggle to cover all possible edge cases

  • Formal verification addresses this by providing machine-checkable guarantees over specified properties, but traditionally demands substantial manual specificati…

  • Closed-World Resolution Against Tool Hallucination in LLM Agents

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19425v1 Announce Type: new.
      • Abstract: Tool-augmented large language model (LLM) agents fail in a way no tool-selection or tool-security method addresses: they call tools that do not exist and pass arguments no schema declares.
      • Existing defenses either pick the right tool (selection) or constrain what an agent may do with real tools (gating), both of which presuppose the emitted call refers to a real tool at all.
      • We show this is a structural blind spot: a hallucinated call is by construction not a decision any gate made, so no gate can reject it.
    • EN Key Points:
      • arXiv:2609.19425v1 Announce Type: new
      • Abstract: Tool-augmented large language model (LLM) agents fail in a way no tool-selection or tool-security method addresses: they call tools that do not exist…
      • Existing defenses either pick the right tool (selection) or constrain what an agent may do with real tools (gating), both of which presuppose the emitted call r…
      • We show this is a structural blind spot: a hallucinated call is by construction not a decision any gate made, so no gate can reject it

  • The syntax and semantics of goals

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19448v1 Announce Type: new.
      • Abstract: In both cognitive science and computer science, goals are conceptualized as cognitive states that flexibly combine with world knowledge to organize and specify purposeful behavior.

  • In this way, goals are compositional representations whose content relates to rational behavior.

  • We here draw attention to goals as representations and their content because it highlights a parallel with other areas in cognitive science - in particular, the syntax-semantics interface in linguistics and logic - while also foregrounding foundational questions about the expressivity, design, and efficiency of different goal representations.

    • EN Highlights:
      • arXiv:2609.19448v1 Announce Type: new
      • Abstract: In both cognitive science and computer science, goals are conceptualized as cognitive states that flexibly combine with world knowledge to organize an…
      • In this way, goals are compositional representations whose content relates to rational behavior
      • We here draw attention to goals as representations and their content because it highlights a parallel with other areas in cognitive science - in particular, the…

ArXiv cs.CL (B_intro+search) Link to heading

  • Modality Discrepancy Transformer for Ambivalence and Hesitancy Recognition

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.19148v1 Announce Type: new.
      • Abstract: Ambivalence and hesitancy (A/H) are affective states in which individuals express contradictory signals across facial, vocal, and linguistic channels.
      • Automatically recognising A/H in clinical videos requires detecting cross-modal disagreement – the signal that standard fusion methods suppress.
      • Based on the conflict-aware multimodal fusion framework of Bekhouche et al., we present the Modality Discrepancy Transformer (MDT).
    • EN Highlights:
      • arXiv:2609.19148v1 Announce Type: new
      • Abstract: Ambivalence and hesitancy (A/H) are affective states in which individuals express contradictory signals across facial, vocal, and linguistic channels
      • Automatically recognising A/H in clinical videos requires detecting cross-modal disagreement – the signal that standard fusion methods suppress

  • Based on the conflict-aware multimodal fusion framework of Bekhouche et al., we present the Modality Discrepancy Transformer (MDT)

  • Subliminal Prompting Beyond Static Geometry: Causal Depth and Multi-Token Confounds

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19149v1 Announce Type: new.
      • Abstract: Subliminal learning shows that language models can transmit a hidden trait through outputs that appear unrelated to it.
      • One proposed explanation, token entanglement, links animal and number tokens through the model’s output vocabulary.
      • Yet existing measurements answer different questions: whether outputs co-vary, fixed output vectors align, an answer can be read from a hidden state, or that state causally controls the answer.
    • EN Key Points:
      • arXiv:2609.19149v1 Announce Type: new
      • Abstract: Subliminal learning shows that language models can transmit a hidden trait through outputs that appear unrelated to it
      • One proposed explanation, token entanglement, links animal and number tokens through the model’s output vocabulary
      • Yet existing measurements answer different questions: whether outputs co-vary, fixed output vectors align, an answer can be read from a hidden state, or that st…

  • Sampling Reveals Style: Unsupervised, Training-Free Discovery of Prompt-Conditional Stylistic Axes in LLM Activations

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19150v1 Announce Type: new.
      • Abstract: Large language models (LLMs) encode rich stylistic structure in their hidden activations, but discovering which stylistic dimensions are salient for a given prompt typically requires supervised contrastive data.

  • We present a training-free, prompt-conditional alternative: we repeatedly sample completions of a single prompt at elevated temperature, apply Principal Component Analysis (PCA) to the pooled hidden activations, and label the resulting axes automatically from the pole generations.

    • We validate the discovered axes against 245 human-elicited stylistic annotations in a two-phase study.
    • EN Highlights:
      • arXiv:2609.19150v1 Announce Type: new
      • Abstract: Large language models (LLMs) encode rich stylistic structure in their hidden activations, but discovering which stylistic dimensions are salient for a…
      • We present a training-free, prompt-conditional alternative: we repeatedly sample completions of a single prompt at elevated temperature, apply Principal Compone…
      • We validate the discovered axes against 245 human-elicited stylistic annotations in a two-phase study

  • What Users Think of Generative AI: A Cross-Platform NLP Analysis of Trust and Friction in App Store Reviews

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.19151v1 Announce Type: new.
      • Abstract: Generative AI (GenAI) applications have achieved rapid consumer adoption, yet little large-scale research examines user-perceived quality, trust, and adoption barriers.
      • We present one of the first cross-application analyses of app store reviews for six major GenAI applications (ChatGPT, Gemini, Microsoft Copilot, Claude, DeepSeek, and Perplexity), comprising 17,012 English-language reviews from Google Play and the Apple App Store.
      • We combine BERTopic topic modeling with RoBERTa sentiment classification and evaluate cross-application differences using chi-square, Kruskal-Wallis, and multinomial logistic regression with Bonferroni correction.
    • EN Highlights:
      • arXiv:2609.19151v1 Announce Type: new

  • Abstract: Generative AI (GenAI) applications have achieved rapid consumer adoption, yet little large-scale research examines user-perceived quality, trust, and…

  • We present one of the first cross-application analyses of app store reviews for six major GenAI applications (ChatGPT, Gemini, Microsoft Copilot, Claude, DeepSe…

  • We combine BERTopic topic modeling with RoBERTa sentiment classification and evaluate cross-application differences using chi-square, Kruskal-Wallis, and multin…

  • FakeSpotter: A content and strategy agnostic Viral Misinformation Detection Tool

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: - arXiv:2609.19152v1 Announce Type: new.
      • Abstract: Misinformation detection tools often rely on binary true and false classifications or models trained on historical examples, limiting their usefulness when novel misleading narratives emerge.
      • Here, we present FakeSpotter, a content- and strategy-agnostic tool designed to estimate the viral misinformation risk of textual content by measuring structural fingerprints of misinformation rather than directly adjudicating truthfulness.
      • FakeSpotter operationalizes a theory-driven framework across linguistic, narrative, logical, and critical-thinking dimensions, using repeated LLM assessments and domain-specific logistic regression classifiers for short and long texts.
    • EN Key Points:
      • arXiv:2609.19152v1 Announce Type: new
      • Abstract: Misinformation detection tools often rely on binary true and false classifications or models trained on historical examples, limiting their usefulness…
      • Here, we present FakeSpotter, a content- and strategy-agnostic tool designed to estimate the viral misinformation risk of textual content by measuring structura…
      • FakeSpotter operationalizes a theory-driven framework across linguistic, narrative, logical, and critical-thinking dimensions, using repeated LLM assessments an…

  • Stop Removing Stopwords: How an Inherited Preprocessing Default Distorts Legal Text-as-Data

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.19153v1 Announce Type: new.
      • Abstract: Empirical legal scholarship increasingly treats judicial text as data, and much of it still runs on sparse, interpretable pipelines – TF-IDF features and linear classifiers – because the textual feature is often the object of study, not merely a means to a prediction.
      • Yet these pipelines inherit a chain of preprocessing defaults from mid-century information retrieval that were never validated against classification accuracy, the most entrenched being stopword removal.
      • This study introduces an exhaustive single-word ablation that measures a preprocessing step’s effect directly against the downstream objective, and applies it to stopword removal as the hardest case to dislodge.
    • EN Highlights:
      • arXiv:2609.19153v1 Announce Type: new
      • Abstract: Empirical legal scholarship increasingly treats judicial text as data, and much of it still runs on sparse, interpretable pipelines – TF-IDF features…
      • Yet these pipelines inherit a chain of preprocessing defaults from mid-century information retrieval that were never validated against classification accuracy,…
      • This study introduces an exhaustive single-word ablation that measures a preprocessing step’s effect directly against the downstream objective, and applies it t…

  • Neo-Classic: A Benchmark for Evaluating Linguistic-Aesthetic Reasoning in Classical Chinese Poetry

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.19154v1 Announce Type: new.
      • Abstract: While Large Language Models (LLMs) achieve high accuracy on established Classical Chinese Poetry benchmarks, it remains challenging to distinguish transferable Linguistic-Aesthetic Reasoning from reliance on familiar pre-training patterns.

  • To address this issue, we introduce Neo-Classic, an evaluation benchmark that combines a constructionist Out-of-Sample (OOS) dataset with a suite of reverse understanding probes.

    • Unlike traditional benchmarks that rely on verification or generation over historical corpora, Neo-Classic comprises strictly metrical poetry authored by contemporary experts, reducing the possibility of direct retrieval.
    • EN Key Points:
      • arXiv:2609.19154v1 Announce Type: new
      • Abstract: While Large Language Models (LLMs) achieve high accuracy on established Classical Chinese Poetry benchmarks, it remains challenging to distinguish tra…
      • To address this issue, we introduce Neo-Classic, an evaluation benchmark that combines a constructionist Out-of-Sample (OOS) dataset with a suite of reverse und…
      • Unlike traditional benchmarks that rely on verification or generation over historical corpora, Neo-Classic comprises strictly metrical poetry authored by contem…

  • Towards Proactive Detection of User-Side Implicit Conflicts in Human-LLM Dialogue

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.19155v1 Announce Type: new.
      • Abstract: In Human-LLM dialogue, follow-up user utterances may implicitly conflict with earlier intents, leading the LLM to misinterpret user needs and generate inappropriate responses.
      • A reliable dialogue system should proactively detect user-side conflicts before generating a response and seek clarification when necessary.
      • However, prior work has largely focused on LLM-side conflicts, leaving user-side conflicts underexplored.
    • EN Key Points:
      • arXiv:2609.19155v1 Announce Type: new
      • Abstract: In Human-LLM dialogue, follow-up user utterances may implicitly conflict with earlier intents, leading the LLM to misinterpret user needs and generate…
      • A reliable dialogue system should proactively detect user-side conflicts before generating a response and seek clarification when necessary.

  • However, prior work has largely focused on LLM-side conflicts, leaving user-side conflicts underexplored

  • Reflective Recovery: A Self-Supervised Method for Reasoning by Learning from Mistakes

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: arXiv:2609.19156v1 Announce Type: new.
      • Abstract: Data-driven fine-tuning is widely adopted to enhance reasoning in Large Language Models (LLMs) due to its simplicity and efficiency.
      • However, mainstream imitation learning methods that rely exclusively on perfect reasoning trajectories suffer from a Scaling Collapse: when the problem set is limited, increasing positive examples fails to yield continuous improvement.
      • However, during inference, an LLM can not guarantee that every intermediate step is correct and is therefore prone to errors.
    • EN Key Points:
      • arXiv:2609.19156v1 Announce Type: new
      • Abstract: Data-driven fine-tuning is widely adopted to enhance reasoning in Large Language Models (LLMs) due to its simplicity and efficiency
      • However, mainstream imitation learning methods that rely exclusively on perfect reasoning trajectories suffer from a Scaling Collapse: when the problem set is l…
      • However, during inference, an LLM can not guarantee that every intermediate step is correct and is therefore prone to errors

  • VisKG-LM: Compiling Knowledge Graphs into Visual Memory for Multiple-Choice Question Answering

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: arXiv:2609.19158v1 Announce Type: new.
      • Abstract: Knowledge graphs are usually integrated into question answering by encoding a retrieved subgraph with a graph neural network and fusing it with the language model in the online inference path.
      • The same subgraph is therefore re-encoded from scratch every time a pair is scored, across training epochs, seeds, and evaluation runs, even though the knowledge graph never changes.

  • We ask whether the retrieved knowledge graphs can instead be compiled once, offline, and then accessed as read-only memory.

    • EN Highlights:
      • arXiv:2609.19158v1 Announce Type: new
      • Abstract: Knowledge graphs are usually integrated into question answering by encoding a retrieved subgraph with a graph neural network and fusing it with the la…
      • The same subgraph is therefore re-encoded from scratch every time a pair is scored, across training epochs, seeds, and evaluation runs, even though the knowledg…
      • We ask whether the retrieved knowledge graphs can instead be compiled once, offline, and then accessed as read-only memory

ArXiv cs.LG (B_intro+search) Link to heading

  • Generative Query Suggestion via Intent Coverage and Query-Level Credit Assignment

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19209v1 Announce Type: new.
      • Abstract: Generative query suggestion aims to enhance user engagement by anticipating user intents and recommending relevant follow-up queries.
      • A central challenge is to generate slates whose individual queries are useful while the slate covers distinct intents.
      • We propose an Intent-Driven Query Suggestion Framework with dual-stage optimization.
    • EN Highlights:
      • arXiv:2609.19209v1 Announce Type: new
      • Abstract: Generative query suggestion aims to enhance user engagement by anticipating user intents and recommending relevant follow-up queries
      • A central challenge is to generate slates whose individual queries are useful while the slate covers distinct intents
      • We propose an Intent-Driven Query Suggestion Framework with dual-stage optimization

  • Layer-wise Curriculum Learning for Efficient LLM Compression

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19213v1 Announce Type: new.
      • Abstract: In this paper, we introduce layer-wise curriculum learning for efficient LLM compression.

  • The proposed method facilitates the knowledge transfer from the teacher model to the student model, utilizing a curriculum learning approach that begins with easier optimization tasks and progressively tackles harder ones.

  • In order to adopt the layer-wise learning in LLM compression, we partition the whole model into multiple segments consisting of layers, thereby enabling more computationally efficient knowledge transfer for LLMs.

    • EN Key Points:
      • arXiv:2609.19213v1 Announce Type: new
      • Abstract: In this paper, we introduce layer-wise curriculum learning for efficient LLM compression
      • The proposed method facilitates the knowledge transfer from the teacher model to the student model, utilizing a curriculum learning approach that begins with ea…
      • In order to adopt the layer-wise learning in LLM compression, we partition the whole model into multiple segments consisting of layers, thereby enabling more co…

  • Block Parallelism For Efficient Distributed Long-Context Diffusion Language Model Training

    • Release Time: 2026-09-18 12:00 Beijing Time
    • Abstract: - arXiv:2609.19242v1 Announce Type: new.
      • Abstract: Block diffusion language models (BDLMs) combine autoregressive dependencies across blocks with parallel denoising within blocks, but long-context training is constrained by distributed attention communication and activation memory.
      • Conventional context parallelism (CP) shards the combined clean-plus-corrupted sequence by position, communicating shared clean K/V together with block-specific corrupted K/V and their gradients.
      • We observe that the BDLM objective separates over target blocks.
    • EN Key Points:
      • arXiv:2609.19242v1 Announce Type: new
      • Abstract: Block diffusion language models (BDLMs) combine autoregressive dependencies across blocks with parallel denoising within blocks, but long-context trai…

  • Conventional context parallelism (CP) shards the combined clean-plus-corrupted sequence by position, communicating shared clean K/V together with block-specific…

  • We observe that the BDLM objective separates over target blocks

  • Randomized SVD Approximations for Spectral Co-Clustering of Word-Document Matrices

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Pending Translation] - arXiv:2609.19243v1 Announce Type: new.
      • Abstract: Spectral co-clustering is a useful tool for discovering latent structure in word-document matrices, but its reliance on singular value decomposition (SVD) can make standard formulations expensive on high-dimensional data.
      • This paper presents two randomized approximations for normalized spectral co-clustering of bipartite text data when the numbers of document and word clusters may differ.
      • The first method uses randomized SVD through random projection, while the second combines partial SVD with element-wise random sampling.
    • EN Key Points:
      • arXiv:2609.19243v1 Announce Type: new
      • Abstract: Spectral co-clustering is a useful tool for discovering latent structure in word-document matrices, but its reliance on singular value decomposition (…
      • This paper presents two randomized approximations for normalized spectral co-clustering of bipartite text data when the numbers of document and word clusters ma…
      • The first method uses randomized SVD through random projection, while the second combines partial SVD with element-wise random sampling

  • Radio-Frequency Convolutional Neural Networks

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [Pending Translation] - arXiv:2609.19279v1 Announce Type: new.
      • Abstract: Running artificial intelligence (AI) models directly on edge devices such as smartphones, wearables, and drones offers low latency, pervasive scalability, and data privacy, but these devices rarely carry the computing capability that modern neural networks demand.

  • Edge accelerators have been developed in response, yet each adds computing hardware to devices already constrained in size, weight, power, and cost (SWaP-C).

  • An alternative lies in what these devices already carry: the frequency mixer in every wireless radio multiplies signals in time, natively performing convolution in the frequency domain.

    • EN Key Points:
      • arXiv:2609.19279v1 Announce Type: new
      • Abstract: Running artificial intelligence (AI) models directly on edge devices such as smartphones, wearables, and drones offers low latency, pervasive scalabil…
      • Edge accelerators have been developed in response, yet each adds computing hardware to devices already constrained in size, weight, power, and cost (SWaP-C)
      • An alternative lies in what these devices already carry: the frequency mixer in every wireless radio multiplies signals in time, natively performing convolution…

  • Learning-Induced Dynamical Transition in Recurrent Neural Networks

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.19288v1 Announce Type: new.
      • Abstract: Learning in recurrent neural networks can fundamentally reshape their underlying dynamics, transforming initially chaotic activity into stable task-dependent behavior.
      • We develop a non-equilibrium dynamical mean-field theory(DMFT) to describe this transition during learning.
      • We show that a slow feedback-driven learning process generates an evolving effective feedback strength that drives the network through a transition from chaotic to stable dynamics defined by a bifurcation of the DMFT solution.
    • EN Key Points:
      • arXiv:2609.19288v1 Announce Type: new
      • Abstract: Learning in recurrent neural networks can fundamentally reshape their underlying dynamics, transforming initially chaotic activity into stable task-de…
      • We develop a non-equilibrium dynamical mean-field theory(DMFT) to describe this transition during learning

  • We show that a slow feedback-driven learning process generates an evolving effective feedback strength that drives the network through a transition from chaotic…

  • Personalized Federated Hierarchical Gaussian Processes for Privacy-Preserving Modeling of Heterogeneous Distributed Systems

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Summary: [Translation Pending] - arXiv:2609.19337v1 Announce Type: new.
      • Abstract: We present Personalized Federated Hierarchical Gaussian Processes (pFedHGP) for probabilistic regression and classification when data are distributed across heterogeneous clients.
      • Each client’s latent function decomposes into (i) a shared global component, (ii) a client-specific deviation that shares the global kernel structure, and (iii) a flexible local residual.
      • Sparse inducing-variable approximations and federated variational inference keep raw data local while the server synchronizes only low-dimensional statistics for the shared component.
    • EN Key Points:
      • arXiv:2609.19337v1 Announce Type: new
      • Abstract: We present Personalized Federated Hierarchical Gaussian Processes (pFedHGP) for probabilistic regression and classification when data are distributed…
      • Each client’s latent function decomposes into (i) a shared global component, (ii) a client-specific deviation that shares the global kernel structure, and (iii)…
      • Sparse inducing-variable approximations and federated variational inference keep raw data local while the server synchronizes only low-dimensional statistics fo…

  • How to Guide Your Language Flow

    • Publication Time: 2026-09-18 12:00 Beijing Time
    • Summary: [Translation Pending] - arXiv:2609.19356v1 Announce Type: new.
      • Abstract: We introduce a new method to guide flow matching models.
      • Our approach, which we call probe guidance, uses the frozen internal states of an existing diffusion model to construct a guidance signal.

  • This works using a similar principle as autoguidance, but eliminates the need for an additional forward pass at inference time and provides a reliable path to ensure that the weak and strong model share similar dynamics.

    • EN Highlights:
      • arXiv:2609.19356v1 Announce Type: new
      • Abstract: We introduce a new method to guide flow matching models
      • Our approach, which we call probe guidance, uses the frozen internal states of an existing diffusion model to construct a guidance signal
      • This works using a similar principle as autoguidance, but eliminates the need for an additional forward pass at inference time and provides a reliable path to e…

  • Smart Insole Human Activity Recognition for Continuous Monitoring in Elderly Care

    • Published: 2026-09-18 12:00 Beijing Time
    • Abstract: [Pending Translation] - arXiv:2609.19359v1 Announce Type: new.
      • Abstract: Falls in older adults are often preceded by changes in mobility, balance, and postural transitions.
      • This paper presents a wireless smart insole platform and machine-learning workflow for recognizing sitting, standing, walking, and unstable walking from plantar-pressure and inertial signals.
      • Each insole integrates 16 active pressure-sensing locations and a six-dimensional IMU stream consisting of tri-axial acceleration and angular velocity.
    • EN Highlights:
      • arXiv:2609.19359v1 Announce Type: new
      • Abstract: Falls in older adults are often preceded by changes in mobility, balance, and postural transitions
      • This paper presents a wireless smart insole platform and machine-learning workflow for recognizing sitting, standing, walking, and unstable walking from plantar…
      • Each insole integrates 16 active pressure-sensing locations and a six-dimensional IMU stream consisting of tri-axial acceleration and angular velocity

  • Stiefel Attention: When the Geometry of Transformer Projection Matrices Dominates Optimizer Choice—and When It Does Not

  • Published: 2026-09-18 12:00 Beijing Time

  • Summary: [To be translated] - arXiv:2609.19363v1 Announce Type: new.

    • Abstract: The query and key projections $\WQ,\WK$ in attention are almost always trained by Euclidean optimizers with no constraint on their geometry.
    • We constrain them to the Stiefel manifold and optimize them there with a Riemannian Adam that carries one scalar second moment per frame, caps its step by a trust region, and retracts polarly.
    • Four propositions prove this update is steepest descent in the embedded metric, independent of gradient scale, well conditioned, and exactly $\mathrm{O}(d)$-equivariant, each certified numerically in \texttt{float64}.

  • EN Key Points:

    • arXiv:2609.19363v1 Announce Type: new
    • Abstract: The query and key projections $\WQ,\WK$ in attention are almost always trained by Euclidean optimizers with no constraint on their geometry
    • We constrain them to the Stiefel manifold and optimize them there with a Riemannian Adam that carries one scalar second moment per frame, caps its step by a tru…
    • Four propositions prove this update is steepest descent in the embedded metric, independent of gradient scale, well conditioned, and exactly $\mathrm{O}(d)$-equ…