System translated (Gemini)

🤖 AI 速览

Today’s focus shifts from model size to system trustworthiness: Fyxer and Perplexity demonstrate agents’ ability for continuous execution in email, software, and production systems, while Occamy-1.0 explores more efficient collaborative open models. Meanwhile, several studies indicate …
📋 文章元数据
发布时间
2026-09-15
类型
ai-daily
字数
6884
阅读时长
33 min

2026-09-15 AI Daily | When Language Models Meet the Real World: The Next Challenge for AI Implementation is Verifiable Quantification Link to heading

Today’s focus shifts from model scale to system trustworthiness: Fyxer and Perplexity demonstrate the continuous execution capabilities of agents in email, software, and production systems, while Occamy-1.0 explores more efficient collaborative open models. Meanwhile, several studies point out that tool orchestration, data traceability, reproducibility, and quantitative reasoning are becoming key factors in measuring whether AI can truly be implemented in high-risk scenarios.

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

The first main thread most worthy of attention today is the agent’s evolution from “model invocation” to “trustworthy working system”: Fyxer’s AI administrative assistant focuses on long-term context across emails, meetings, and applications, while Occamy-1.0 showcases open models for complex collaborative processes. At the same time, research on model capabilities and Agent harness reminds engineering teams to re-evaluate the true contributions of tools, prompts, and control flow. The second main thread is reliability and verifiability: from R2VC’s Retrieve-Verify-Confidence Calibration, to draft-verify-revise for handling instruction ambiguity, and then to quantitative reasoning and event prediction, the core issue is how to make models “prove it” more and “sound like it” less in high-risk scenarios. In addition, discussions on frontier governance and infrastructure are also worth noting: from Dario Amodei’s debate on the pace of AI development, to the vertical integration of computing power, data centers, and software stacks, the capabilities race is simultaneously evolving into a competition of governance and industrial organizational capabilities.

🌐 X Platform AI Hot News Link to heading

Topic 1: Anthropic Launches Claude Mods for Custom Terminal Plugins Link to heading

  • Category: AI · News
  • Overview: Hot for: 3 hours ago, Related Posts: 335
  • What happened: Anthropic launched a feature called Claude Mods for creating and customizing terminal plugins.
  • Why it’s important: This indicates that AI assistants are expanding from standalone conversational tools to platforms embeddable in developer terminal workflows, which helps enhance automation capabilities and ecosystem expansion.
  • Discussion overview: Discussion on X is expected to focus on the scope of plugin customization, development experience, compatibility with existing terminal tools, as well as issues such as security, permissions, and maintenance costs.

Topic 2: Yann LeCun Mocks 2019 GPT-2 Release Fears Link to heading

  • Category: AI · Other
  • Overview: Hot for: 1 day ago, Related Posts: 5800
  • What happened: Yann LeCun looked back and mocked the concerns surrounding potential risks and impacts when GPT-2 was released in 2019.
  • Why it’s important: This reflects the industry’s continuous introspection on generative AI risk assessment, the speed of technological progress, and open release strategies.
  • Discussion overview: Discussion on X mainly focuses on whether the judgment of GPT-2’s capabilities back then was overly cautious, whether subsequent model developments proved these concerns, and how AI companies should balance open sharing and security risks.

Topic 3: Meta’s Muse AI Agent Wins Early Praise for Everyday Help Link to heading

  • Category: AI · News
  • Overview: Hot for: 2 days ago, Related Posts: 1900
  • What happened: Meta’s Muse AI Agent received early praise for its practical performance in daily tasks.
  • Why it’s important: This indicates that AI Agents are moving beyond mere Q&A to an application form capable of assisting with real-world daily affairs, which helps test their practical implementation value.
  • Discussion overview: Discussion on X mainly focuses on its convenience, task completion effectiveness, and user experience, while also raising questions about reliability, privacy, security, and whether early publicity was exaggerated.

Topic 4: AI Speeds Product Building but Demands Deeper User Insight Link to heading

  • Category: AI · News
  • Overview: Hot for: 7 hours ago, Related Posts: 804
  • What happened: Discussion points out that AI is accelerating product building, but product success still depends on a deeper understanding of user needs.
  • Why it’s important: This indicates that AI can significantly improve development and iteration efficiency but cannot replace user research, demand judgment, and product decisions, reflecting that AI applications are shifting from simple efficiency improvement to human-computer collaboration.
  • Discussion overview: Discussion on X mainly revolves around the contradiction between AI-driven development acceleration and insufficient user insight: one side emphasizes that AI lowers the threshold for product building, while the other side worries that rapid output does not equate to solving real needs.

Topic 5: Meta’s Muse AI Agent Tops Download Charts Days After Launch Link to heading

  • Category: AI · News
  • Overview: Hot for: 21 hours ago, Related Posts: 2100
  • What happened: Meta’s Muse AI Agent topped the download charts days after its launch, showing its rapid attention in the consumer market.
  • Why it matters: This indicates that AI applications are shifting from one-off Q&A to intelligent agents capable of continuous task execution, which helps in observing real user demand for agent-based interactions and productized AI.
  • Discussion overview: Discussions on X are centered on the idea that “loops are the new prompts,” where continuous task iteration replaces single prompts. Meanwhile, some question whether download volume can represent actual usage and effectiveness, and are focusing on the accuracy, privacy, and risks of autonomous actions by agents.

Topic 6: Community Ranks Top Multi-Agent Harnesses for AI Development Link to heading

  • Category: AI · News
  • Overview: Trending 13 hours ago, 66 related posts
  • What it is: Users on the X platform are ranking and discussing the most suitable multi-agent runtime frameworks for AI development, linking them to issues like large-scale AI deployment, open weights, and governance.
  • Why it matters: Multi-agent frameworks determine how multiple AI agents collaborate, invoke tools, and complete complex development tasks. They are the essential infrastructure for transitioning AI from single-model applications to engineered, large-scale deployment.
  • Discussion overview: Current discussions focus on the differences among various multi-agent frameworks regarding development efficiency, reliability, scalability, and security governance. Some perspectives are centered on their commercial and defense application prospects, while others are concerned about the risks posed by open weights, loss of agent control, and regulatory gaps.

Summary of AI Public Opinion on X Today Link to heading

The main theme of today’s public opinion is that AI is rapidly evolving from a conversational tool into an engineered platform embedded in terminals, capable of continuous task execution and supporting multi-agent collaboration. Discussions around Claude Mods, Muse Agent, and related frameworks collectively reflect this trend. The general consensus is that AI significantly lowers the barriers to product development and automation and is starting to demonstrate its value in handling real-world tasks. However, product success still hinges on genuine user needs, reliable user experience, and continuous iteration, rather than solely on model capabilities or download figures. Points of disagreement mainly center on whether early risk assessments (like those for GPT-2) were excessive, how far open-sourcing weights and open releases should go, and whether the hype surrounding agent-based products can translate into stable, real-world effectiveness. Potential risks include permission controls for plugins and agents, privacy leaks, task failures, loss of autonomous control, and governance gaps that could arise from the large-scale deployment of multi-agent systems in commercial and defense applications.

💡 Influencer Insights Link to heading

No influencer insights for today. We recommend reading the in-depth content on the Watch List.

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

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

Acquired.fm (A_full) Link to heading

  • The Home Depot
    • Release Time: 2026-09-14 11:35 Beijing Time
    • Abstract: [Awaiting Translation] - The Home Depot’s founding story is like an Avengers movie… if the Avengers got fired, went broke, and stacked empty paint cans ten feet high to look legitimate.
      • After being unceremoniously fired from their previous hardware chain at ages 48 and 35, Bernie Marcus and Arthur Blank took the words of their New York banker Ken Langone (who had also just accidentally caused their firings) to heart: they’d just been “kicked in the ass with a golden horseshoe.” They proceeded to author the greatest compounding story in American retail history, helped by some legendary cameos along the way from Sol Price, Jamie Dimon, and Ross Perot (to name a few).
      • And the ending is as good as any superhero film: from its 1981 IPO to today, Home Depot has been the single highest-returning equity in the entire US stock market — higher than Apple, Microsoft, Berkshire Hathaway, and everything else
  • Many thanks to our fantastic Fall ‘26 Season partners:.
      • Sign up for email updates , get our takeaways and research photos from each episode, and vote on future topics!
    • EN Key Points:
      • The Home Depot’s founding story is like an Avengers movie… if the Avengers got fired, went broke, and stacked empty paint cans ten feet high to look legitimate
      • After being unceremoniously fired from their previous hardware chain at ages 48 and 35, Bernie Marcus and Arthur Blank took the words of their New York banker K…
      • And the ending is as good as any superhero film: from its 1981 IPO to today, Home Depot has been the single highest-returning equity in the entire US stock mark…
      • Many thanks to our fantastic Fall ‘26 Season partners:

All-In Podcast (A_full) Link to heading

  • Jensen Huang: The Doomer Hoax, Superintelligence Is Here, and The Future of AI (ft. President Trump)
    • Published: 2026-09-15 05:24 Beijing Time
    • Summary: [Pending Translation] - 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) Jensen Huang joins The Besties
      • (1:39) Thoughts on Dario’s blog, Frontier Labs calling to slow down AI, and Doomer psychology
      • (9:58) Sensible AI regulation and RSI
      • (16:05) Hugging Face acquisition, future of Open Source, and the race with China

Stratechery by Ben Thompson (A_full) Link to heading

  • Pacing the Frontier, AI’s Digital Limits, AI Commissars
    • Publication Time: 2026-09-14 18:00 Beijing Time
    • Summary: Dario Amodei wants to pace the frontier; it’s an unrealistic proposal that seems mostly geared to political control of AI.
      • $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:
      • Dario Amodei wants to pace the frontier; it’s an unrealistic proposal that seems mostly geared to political control of AI.

OpenAI Blog (A_full) Link to heading

  • How Fyxer built an AI executive assistant people trust

    • Publication Time: 2026-09-14 20:00 Beijing Time
    • Summary: For many professionals, work means keeping track of conversations and commitments across inboxes, meetings, messages, and apps.
      • Without rock-solid context, those commitments can fall through the cracks, damaging projects and relationships.
      • Fyxer built an AI executive assistant that follows the thread as work moves across tools.
      • It combines the latest OpenAI models with more than 500,000 hours of executive assistant workflows, dividing the work among dozens of specialized models that improve through real user feedback.
      • Email is one of the clearest places to see it in action.
    • EN Key Points:
      • Fyxer uses OpenAI models, fine-tuning, memory, and real user feedback to organize inboxes and draft emails in each user’s voice.

  • Perplexity trusts GPT-6 Astra with end-to-end systems

    • Publication Time: 2026-09-14 08:00 Beijing Time
    • Summary: Perplexity uses Astra to write communications, change software, and monitor production systems, and checks in much less frequently than with earlier models.
      • As an AI-powered answer engine, Perplexity is deeply focused on search and accuracy.

  • Its ability to process large amounts of information is critically important.

  • Johnny Ho, Cofounder and Chief Strategy Officer, observes that every time the model gets better at writing code, Perplexity’s search engine improves too.

  • It becomes able to write better programs that search the web and internal information and summarize it very concisely.

    • EN Key Points:
      • Perplexity uses Astra to write communications, change software, and monitor production systems, and checks in much less frequently than with earlier models.

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

  • Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work

    • Release Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11977v1 Announce Type: new.
      • Abstract: Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations.
      • Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability is delivered.
      • Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning.
    • EN Key Points:
      • arXiv:2609.11977v1 Announce Type: new
      • Abstract: Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations
      • Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability…
      • Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning

  • Harness or Model? Isolating the Harness Effect in Agentic Coding with a Contamination-Controlled Private Suite

    • Release Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11987v1 Announce Type: new.

  • A Tale of Two Harnesses: Open-Source vs. Vendor-Native Agentic Coding Systems

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11987v1 Announce Type: new.
      • Abstract: An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software engineer.
      • Vendors ship harnesses tuned to their own models, and practitioners assume the vendor-native pairing solves more tasks.
      • We measure that assumption with paired same-model contrasts on a private, contamination-controlled suite of 256 repository and post-cutoff contest tasks.
    • EN Highlights:
      • arXiv:2609.11987v1 Announce Type: new
      • Abstract: An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software…
      • Vendors ship harnesses tuned to their own models, and practitioners assume the vendor-native pairing solves more tasks
      • We measure that assumption with paired same-model contrasts on a private, contamination-controlled suite of 256 repository and post-cutoff contest tasks

  • Reading the Whole Heart: Latent-Attention Masked Autoencoders for Multimodal Cardiac Representation Learning

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12035v1 Announce Type: new.
      • Abstract: Cardiovascular diagnosis rests on integrating complementary modalities, like ECG, echocardiography, chest radiographs, and clinical variables, each capturing distinct but correlated aspects of cardiac physiology.
      • Yet most medical foundation models remain modality-specific, combining modalities only for finetuning or post-training.
      • This discards the cross-modal evidence clinicians naturally integrate and ignores the structure within each modality.
    • EN Highlights:
      • arXiv:2609.12035v1 Announce Type: new
      • Abstract: Cardiovascular diagnosis rests on integrating complementary modalities, like ECG, echocardiography, chest radiographs, and clinical variables, each ca…

  • Yet most medical foundation models remain modality-specific, combining modalities only for finetuning or post-training

    • This discards the cross-modal evidence clinicians naturally integrate and ignores the structure within each modality

  • Competence-Gated Pooling of Language Models and Priors for Event Forecasting

    • Release Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12101v1 Announce Type: new.
      • Abstract: In hybrid forecasting, a language model is often one of several available signals.
      • A system may already have a market, crowd, or statistical forecast and must decide whether the model adds useful information or should be ignored.
      • The relevant target is therefore not standalone model accuracy, but relative competence, defined as the model’s marginal value beyond the available external forecast.
    • EN Highlights:
      • arXiv:2609.12101v1 Announce Type: new
      • Abstract: In hybrid forecasting, a language model is often one of several available signals
      • A system may already have a market, crowd, or statistical forecast and must decide whether the model adds useful information or should be ignored
      • The relevant target is therefore not standalone model accuracy, but relative competence, defined as the model’s marginal value beyond the available external for…

  • Language Is an Insufficient Substrate for Quantitative Reasoning, and Consequential Domains Need Large Quantitative Models

    • Release Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12105v1 Announce Type: new.
      • Abstract: The prevailing assumption in applied machine learning is that progress on consequential quantitative decisions such as pricing risk, allocating capital, triaging patients, or containing a network intrusion will follow from progress in large language models (LLMs).

  • A language model is trained on a representation of the world that was produced by human description; description is a lossy encoding of the quantitative record, and the loss is irreversible: no downstream model, at any scale, can recover from a description what the description did not encode.

  • We formalize this as a property of the representation on which a model is trained rather than of the model capacity, and we identify three further properties that consequential settings demand of a model and that a language substrate cannot supply by construction: reproducibility, lineage from every output back to the source records that produced.

    • EN Highlights:
      • arXiv:2609.12105v1 Announce Type: new
      • Abstract: The prevailing assumption in applied machine learning is that progress on consequential quantitative decisions such as pricing risk, allocating capita…
      • A language model is trained on a representation of the world that was produced by human description; description is a lossy encoding of the quantitative record,…
      • We formalize this as a property of the representation on which a model is trained rather than of the model capacity, and we identify three further properties th…

  • DU-NO: A Parameter-Efficient Double U-Shaped Neural Operator for Phase-Resolving Wave Modeling

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract: [Translation pending] - arXiv:2609.12115v1 Announce Type: new.
      • Abstract: Phase-resolving wave models such as FUNWAVE-TVD are the accuracy standard for nearshore dynamics, resolving the shoaling, refraction, and breaking of individual waves, but their cost rules them out for the ensembles, uncertainty quantification, and real-time warning that operational forecasting demands.

  • Neural operators promise solver-level accuracy at a fraction of that cost, yet on wave-dominated fields the accurate ones are large: hybrid spectral-convolutional operators such as U-FNO (the strongest baseline in our study after DU-NO) buy their fidelity with tens of millions of parameters.

  • We introduce DU-NO (Double U-shaped Neural Operator), a multiscale U-shaped spectral operator that attaches lightweight convolutional U-Net branches only at its two shallowest encoder and decoder levels.

  • EN Highlights:

    • arXiv:2609.12115v1 Announce Type: new
    • Abstract: Phase-resolving wave models such as FUNWAVE-TVD are the accuracy standard for nearshore dynamics, resolving the shoaling, refraction, and breaking of…
    • Neural operators promise solver-level accuracy at a fraction of that cost, yet on wave-dominated fields the accurate ones are large: hybrid spectral-convolution…
    • We introduce DU-NO (Double U-shaped Neural Operator), a multiscale U-shaped spectral operator that attaches lightweight convolutional U-Net branches only at its…

  • When Successful Knowledge Graph Edits Displace Correct Answers: Rank-Level Locality beyond Parameter Support

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract:
      • Editing a knowledge graph embedding (KGE) model to promote a desired answer can displace correct answers from the returned list.
      • Locality tests based only on facts that reuse the edited parameter can miss this ranking effect.
      • We introduce a common rank-displacement audit at three scopes: facts supported by the edited parameter, other correct answers to the target query, and correct answers across queries with the same relation.
    • EN Highlights:
      • arXiv:2609.12116v1 Announce Type: new
      • Abstract: Editing a knowledge graph embedding (KGE) model to promote a desired answer can displace correct answers from the returned list

  • Locality tests based only on facts that reuse the edited parameter can miss this ranking effect

  • We introduce a common rank-displacement audit at three scopes: facts supported by the edited parameter, other correct answers to the target query, and correct a…

  • Mined from Scientific Literature: Process Schemas for Atomic Layer Deposition and Etching in Materials Science

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Summary: 【待翻译】- arXiv:2609.12139v1 Announce Type: new.
      • Abstract: Atomic layer deposition (ALD) and atomic layer etching (ALE) are reported heterogeneously across experimental and simulation literature in materials science, hindering comparison and machine-actionable reuse.
      • We present four domain-expert-reviewed JSON Schemas for ALD and ALE experimental and simulation processes.
      • Curated with schema-miner and grounded in QUDT using schema-miner pro, the schemas structure materials, process conditions, configurations , and measured or predicted results.
    • EN Key Points:
      • arXiv:2609.12139v1 Announce Type: new
      • Abstract: Atomic layer deposition (ALD) and atomic layer etching (ALE) are reported heterogeneously across experimental and simulation literature in materials s…
      • We present four domain-expert-reviewed JSON Schemas for ALD and ALE experimental and simulation processes
      • Curated with schema-miner and grounded in QUDT using schema-miner pro, the schemas structure materials, process conditions, configurations , and measured or pre…

  • Can LLMs in Draft-Verify-Revise Pipelines Resolve Deictic Ambiguity?

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Summary: 【待翻译】- arXiv:2609.12162v1 Announce Type: new.
      • Abstract: Draft-verify-revise is a common LLM orchestration pattern for scaling inference-time compute.
      • One LLM drafts, a second critiques the draft and provides feedback, and a third uses that feedback to revise the draft into the final output.

  • As context cascades between stages, LLMs at different stages can resolve a context-dependent expression such as “previous” differently.

    • EN Highlights:
      • arXiv:2609.12162v1 Announce Type: new
      • Abstract: Draft-verify-revise is a common LLM orchestration pattern for scaling inference-time compute
      • One LLM drafts, a second critiques the draft and provides feedback, and a third uses that feedback to revise the draft into the final output
      • As context cascades between stages, LLMs at different stages can resolve a context-dependent expression such as “previous” differently

  • GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12165v1 Announce Type: new.
      • Abstract: Meeting continuation requires tracking the agenda, speaker roles, participant intentions, and disagreement across long multi-party discussions.
      • We introduce the Meeting Dynamic Forecasting Benchmark (MDFB), constructed from 2,207 real-world meetings and 24,794 future-facing queries.
      • Given a transcript prefix and an active question, a model generates a plausible multi-turn continuation in one call.
    • EN Highlights:
      • arXiv:2609.12165v1 Announce Type: new
      • Abstract: Meeting continuation requires tracking the agenda, speaker roles, participant intentions, and disagreement across long multi-party discussions
      • We introduce the Meeting Dynamic Forecasting Benchmark (MDFB), constructed from 2,207 real-world meetings and 24,794 future-facing queries
      • Given a transcript prefix and an active question, a model generates a plausible multi-turn continuation in one call

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

  • R2VC: Modular Fact-Checking with Retrieval, Verification, and Confidence Calibration

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11955v1 Announce Type: new.

  • Abstract: Large language models are increasingly used for automated fact checking, but end-to-end prompting often entangles evidence retrieval, reasoning, and uncertainty estimation, making failures difficult to diagnose and confidence difficult to trust.

  • We present R2VC, a modular retrieve, reason, verify, calibrate architecture for evidence-grounded fact checking with citations and abstention.

  • R2VC combines hybrid sparse+dense retrieval over Wikipedia, a supervised fine-tuned and DPO-aligned generator that produces diverse structured verdict candidates, an external NLI cross-encoder for evidence-based candidate selection, and a lightweight sequence-level calibrator for confidence estimation and selective abstention.

    • EN Key Points:
      • arXiv:2609.11955v1 Announce Type: new
      • Abstract: Large language models are increasingly used for automated fact checking, but end-to-end prompting often entangles evidence retrieval, reasoning, and u…
      • We present R2VC, a modular retrieve, reason, verify, calibrate architecture for evidence-grounded fact checking with citations and abstention
      • R2VC combines hybrid sparse+dense retrieval over Wikipedia, a supervised fine-tuned and DPO-aligned generator that produces diverse structured verdict candidate…

  • What Counts as a Mistake? Annotating Recitation Events in Quran Memorization Transcripts

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.12085v1 Announce Type: new.
      • Abstract: Checking Quran recitation from an ASR transcript requires distinguishing unresolved mistakes from repetitions, repairs, opening formulas and accepted spelling differences.
      • We report a completed human annotation of 100 production recording cases: 348 scored units and 162 localized events across ten combined labels.
      • An executable evaluator scores labels and word positions together.
    • EN Key Points:
      • arXiv:2609.12085v1 Announce Type: new

  • Abstract: Checking Quran recitation from an ASR transcript requires distinguishing unresolved mistakes from repetitions, repairs, opening formulas and accepted…

  • We report a completed human annotation of 100 production recording cases: 348 scored units and 162 localized events across ten combined labels

  • An executable evaluator scores labels and word positions together

  • Extracting Dataset Mentions in Forced Displacement and FCV Documents: A Weakly Supervised Framework with LLM-Based Label Refinement

    • Release Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.12107v1 Announce Type: new.
      • Abstract: Development and humanitarian organizations produce and support surveys, administrative registries, and other data resources to inform research, policy, and operations, yet systematically identifying where these datasets are referenced remains difficult.
      • Such references are dispersed across research papers, project documents, humanitarian reports, and other unstructured text, limiting both the ability to trace data use and to identify potential gaps in data availability or dissemination.
      • We present a weakly supervised framework for adapting dataset extraction to forced displacement and Fragile, Conflict, and Violence (FCV) documents without first constructing a large manually labeled training corpus.
    • EN Key Points:
      • arXiv:2609.12107v1 Announce Type: new
      • Abstract: Development and humanitarian organizations produce and support surveys, administrative registries, and other data resources to inform research, policy…
      • Such references are dispersed across research papers, project documents, humanitarian reports, and other unstructured text, limiting both the ability to trace d…
      • We present a weakly supervised framework for adapting dataset extraction to forced displacement and Fragile, Conflict, and Violence (FCV) documents without firs…

  • The Cost of Compression: A Rate-Distortion Limit on Factual Hallucination

    • Publish Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12111v1 Announce Type: new.
      • Abstract: Factual hallucination in closed-book question answering is often treated as a coverage problem: a model fails because the relevant fact is absent from its internal memory.
      • This view misses a second source of error.
      • Even when a fact has been observed, finite memory may force it to be stored only approximately.
    • Key Points:
      • arXiv:2609.12111v1 Announce Type: new
      • Abstract: Factual hallucination in closed-book question answering is often treated as a coverage problem: a model fails because the relevant fact is absent from…
      • This view misses a second source of error
      • Even when a fact has been observed, finite memory may force it to be stored only approximately

  • Quantifying Consonant Contributions to Word Intelligibility via Acoustic Masking

    • Publish Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12122v1 Announce Type: new.
      • Abstract: Consonants contribute unequally to whether a word is understood.
      • Given the limited time available for therapy, ranking consonants by contribution to intelligibility helps prioritize intervention targets in motor speech disorders.
      • However, measuring this contribution relies on perceptual studies that are difficult to scale.
    • Key Points:
      • arXiv:2609.12122v1 Announce Type: new
      • Abstract: Consonants contribute unequally to whether a word is understood
      • Given the limited time available for therapy, ranking consonants by contribution to intelligibility helps prioritize intervention targets in motor speech disord…
      • However, measuring this contribution relies on perceptual studies that are difficult to scale

  • Local Edits, Global Ripples: Replay-Informed Policy Adaptation for Workflow Synthesis

    • Publish Time: 2026-09-14 12:00 Beijing Time

  • Summary: [To be translated] - arXiv:2609.12127v1 Announce Type: new.

    • Abstract: Prompt-policy editing offers a practical way to improve agents that synthesize executable workflows without updating the underlying model.
    • However, persistent prompt editing has two coupled properties.
    • First, edit locality does not imply effect locality: an edit confined to one policy segment can ripple through downstream execution, altering behavior beyond the edited segment.
    • EN Key Points:
      • arXiv:2609.12127v1 Announce Type: new
      • Abstract: Prompt-policy editing offers a practical way to improve agents that synthesize executable workflows without updating the underlying model
      • However, persistent prompt editing has two coupled properties
      • First, edit locality does not imply effect locality: an edit confined to one policy segment can ripple through downstream execution, altering behavior beyond th…

  • Population-level measures of perceived food access reveal barriers beyond geographic proximity

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Summary: [To be translated] - arXiv:2609.12132v1 Announce Type: new.
      • Abstract: Food access is multidimensional, but population-level measurement still relies heavily on geography because perceived dimensions of access are difficult to measure at scale.
      • Here, we use 25,125 Google Maps reviews from 49 grocery stores in Raleigh, North Carolina, to measure five dimensions of food access: availability, accessibility, affordability, accommodation, and acceptability.
      • We identify review topics with unsupervised topic modeling and assign them to access dimensions using zero-shot classification, with 85.4% agreement against manual coding.
    • EN Key Points:
      • arXiv:2609.12132v1 Announce Type: new
      • Abstract: Food access is multidimensional, but population-level measurement still relies heavily on geography because perceived dimensions of access are difficu…

  • Here, we use 25,125 Google Maps reviews from 49 grocery stores in Raleigh, North Carolina, to measure five dimensions of food access: availability, accessibilit…

  • We identify review topics with unsupervised topic modeling and assign them to access dimensions using zero-shot classification, with 85.4% agreement against man…

  • GAUGE: When Not to Trust LLM-as-a-Judge in User-Simulated Evaluation of Task-Oriented Agents

    • Posted: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.12191v1 Announce Type: new.
      • Abstract: Comparing and selecting task-oriented LLM agents increasingly relies on a low-cost offline evaluation gate: persona-driven LLM user-simulators converse with each candidate, an LLM-as-a-judge scores the transcripts, and the higher-scoring agent is promoted.
      • We introduce GAUGE, a reusable offline protocol that measures whether this gate’s ranking matches a grounded verifiable reward across 25 agents from six providers on the $\tau^2$-bench and SimulatorArena benchmarks, separating two kinds of evaluation validity that release practices conflate: ranking validity and construct validity.
      • First, a satisfaction-success gap: satisfaction carries essentially no information about task success, as conversations rated satisfied by our blind panel are decorrelated from actual success, with 57.5% of them failing the customer’s task, a pattern consistent across five rater populations, both benchmarks, and every subjective dimension we rated.
    • EN Highlights:
      • arXiv:2609.12191v1 Announce Type: new
      • Abstract: Comparing and selecting task-oriented LLM agents increasingly relies on a low-cost offline evaluation gate: persona-driven LLM user-simulators convers…
      • We introduce GAUGE, a reusable offline protocol that measures whether this gate’s ranking matches a grounded verifiable reward across 25 agents from six provide…

  • First, a satisfaction-success gap: satisfaction carries essentially no information about task success, as conversations rated satisfied by our blind panel are d…

  • Repair Before Reinforce: Context-Augmented Knowledge Graph Reasoning for Multi-Hop Question Answering

    • Release Date: 2026-09-14 12:00 Beijing Time
    • Abstract: [[待翻译]]- arXiv:2609.12230v1 Announce Type: new.
      • Abstract: Question-answering often requires reasoning across multiple connected facts rather than retrieving a single isolated relation.
      • Knowledge graphs (KGs) provide a structured way to represent such facts, but training large language models (LLMs) only on isolated KG head-relation-tail triples may limit their ability to learn the surrounding context needed for multi-hop reasoning.
      • In this work, we propose a context-augmented training framework for multi-hop question-answering.
    • EN Key Points:
      • arXiv:2609.12230v1 Announce Type: new
      • Abstract: Question-answering often requires reasoning across multiple connected facts rather than retrieving a single isolated relation
      • Knowledge graphs (KGs) provide a structured way to represent such facts, but training large language models (LLMs) only on isolated KG head-relation-tail triple…
      • In this work, we propose a context-augmented training framework for multi-hop question-answering

  • Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding

    • Release Date: 2026-09-14 12:00 Beijing Time
    • Abstract: [[待翻译]]- arXiv:2609.12243v1 Announce Type: new.
      • Abstract: Inference-time power sampling via Sequential Monte Carlo (SMC) can substantially improve large language model (LLM) reasoning without requiring post-training.
      • However, many existing SMC approaches rely on equal-weight resampling, which can aggressively prune low-weight trajectories, discarding potentially correct reasoning paths and degrading the genealogical diversity of the search space.

  • To address this, we introduce Chopthin-Consensus Power Sampling (CCPS).

    • EN Highlights:
      • arXiv:2609.12243v1 Announce Type: new
      • Abstract: Inference-time power sampling via Sequential Monte Carlo (SMC) can substantially improve large language model (LLM) reasoning without requiring post-t…
      • However, many existing SMC approaches rely on equal-weight resampling, which can aggressively prune low-weight trajectories, discarding potentially correct reas…
      • To address this, we introduce Chopthin-Consensus Power Sampling (CCPS)

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

  • Fundamental Dynamical Units for Physics-Informed Structural Inference from Perturbation Time-Series in Networked Systems

    • Published: 2026-09-14 12:00 Beijing Time
    • Abstract: [Translation Pending] - arXiv:2609.11934v1 Announce Type: new.
      • Abstract: In networked dynamical systems, the parameter of primary mechanistic interest is signed interaction structure.
      • Recovering this structure from perturbation time-series data is a fundamental identification problem, compounded by three coupled obstacles: the combinatorial complexity of interaction architectures, ambiguity of causal attribution under limited interventions, and state-dependent dynamics that confound structural inference.
      • Each obstacle is structural in origin and calls for a structural solution.
    • EN Highlights:
      • arXiv:2609.11934v1 Announce Type: new
      • Abstract: In networked dynamical systems, the parameter of primary mechanistic interest is signed interaction structure
      • Recovering this structure from perturbation time-series data is a fundamental identification problem, compounded by three coupled obstacles: the combinatorial c…
      • Each obstacle is structural in origin and calls for a structural solution

  • Physics-Informed Conformal Prediction: Embedding PDE Consistency into Distribution-Free Uncertainty Quantification for Neural Operators

    • Published: 2026-09-14 12:00 Beijing Time

  • Summary: [To be translated] - arXiv:2609.11935v1 Announce Type: new.

    • Abstract: Neural operators such as the Fourier Neural Operator (FNO) achieve remarkable accuracy in approximating solutions to partial differential equations (PDEs).
    • However, providing rigorous uncertainty estimates remains an open challenge.
    • We propose Physics-Informed Conformal Prediction (PI-CP), a framework that embeds PDE residuals into the nonconformity score of split conformal prediction, producing prediction intervals that are (i) distribution-free with provable coverage guarantees, and (ii) spatially adaptive when the PDE residual correlates with prediction error – tighter where physics is well-satisfied, wider where it is violated.
    • EN Key Points:
      • arXiv:2609.11935v1 Announce Type: new
      • Abstract: Neural operators such as the Fourier Neural Operator (FNO) achieve remarkable accuracy in approximating solutions to partial differential equations (P…
      • However, providing rigorous uncertainty estimates remains an open challenge
      • We propose Physics-Informed Conformal Prediction (PI-CP), a framework that embeds PDE residuals into the nonconformity score of split conformal prediction, prod…

  • Fed-Equilibrium Framework for Topological Pareto Control in Robust and Fair Clinical Federated Learning

    • Published: 2026-09-14 12:00 Beijing Time
    • Summary: [To be translated] - arXiv:2609.11937v1 Announce Type: new.
      • Abstract: The deployment of Federated Learning (FL) in multi-center clinical networks faces the challenge of “knowledge dominance,” where high-volume hubs naturally overwhelm minority community nodes, implicitly treating the distinct clinical patterns of smaller cohorts as outliers.
      • Existing geometric defenses provide a security baseline but leave this efficiency-fairness dilemma unresolved.
      • To bridge this gap, we propose Fed-Equilibrium, a framework that advances the paradigm from simple defense to topological equilibrium.
    • EN Key Points:

  • arXiv:2609.11937v1 Announce Type: new

    • Abstract: The deployment of Federated Learning (FL) in multi-center clinical networks faces the challenge of “knowledge dominance,” where high-volume hubs natur…
    • Existing geometric defenses provide a security baseline but leave this efficiency-fairness dilemma unresolved
    • To bridge this gap, we propose Fed-Equilibrium, a framework that advances the paradigm from simple defense to topological equilibrium

  • Efficient AI Model Deployment Using Quantization Analysis Tool

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Summary: [Translation Pending] - arXiv:2609.11954v1 Announce Type: new.
      • Abstract: As deep learning models are increasingly deployed on resource constrained devices, the demand for efficient model optimization techniques continues to grow.
      • Effective deployment of AI models on edge and low power platforms requires optimization methods that reduce model size and computational cost while maintaining high accuracy.
      • This paper presents Quantization Analysis Tool, a practical system designed to streamline quantization workflows and support performance efficient model deployment.
    • EN Key Points:
      • arXiv:2609.11954v1 Announce Type: new
      • Abstract: As deep learning models are increasingly deployed on resource constrained devices, the demand for efficient model optimization techniques continues to…
      • Effective deployment of AI models on edge and low power platforms requires optimization methods that reduce model size and computational cost while maintaining…
      • This paper presents Quantization Analysis Tool, a practical system designed to streamline quantization workflows and support performance efficient model deploym…

  • Performance, Efficiency and Collapse – Advantages and Challenges in Offline Post-training of Code LLMs

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Summary: [Translation Pending] - arXiv:2609.11956v1 Announce Type: new.

  • Abstract: Post-training with reinforcement learning (RL) is a critical phase in the development of code-generating large language models (LLMs), as it ensures adherence to instructions and the production of functionally correct code.

  • This process typically requires computationally intensive code sample generation from Transformer-based LLMs and substantial GPU-CPU communication for sequence verification.

  • To address these computational challenges, this work examines whether RL-based post-training can be performed entirely offline by leveraging existing datasets rather than generating new samples.

    • EN Highlights:
      • arXiv:2609.11956v1 Announce Type: new
      • Abstract: Post-training with reinforcement learning (RL) is a critical phase in the development of code-generating large language models (LLMs), as it ensures a…
      • This process typically requires computationally intensive code sample generation from Transformer-based LLMs and substantial GPU-CPU communication for sequence…
      • To address these computational challenges, this work examines whether RL-based post-training can be performed entirely offline by leveraging existing datasets r…

  • Look Before You Leap: Pre-Action Verification for LLM Agents

    • Release Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11957v1 Announce Type: new.
      • Abstract: An LLM agent acts on the world by emitting actions: shell commands to run, edits to apply.
      • A wrong action does not always fail loudly; it can fail silently, producing a plausible but incorrect effect that raises no error.
      • We argue that a cheap deterministic check, run before an action takes effect, is an effective and underused form of agent oversight, and we study it across two action modalities in one framework.
    • EN Highlights:
      • arXiv:2609.11957v1 Announce Type: new
      • Abstract: An LLM agent acts on the world by emitting actions: shell commands to run, edits to apply

  • A wrong action does not always fail loudly; it can fail silently, producing a plausible but incorrect effect that raises no error

  • We argue that a cheap deterministic check, run before an action takes effect, is an effective and underused form of agent oversight, and we study it across two…

  • Decoding Mixture Perception through Computational Modeling of Component Interactions

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11958v1 Announce Type: new.
      • Abstract: Olfaction played an indispensable role throughout human evolution and civilization.
      • Even in the contemporary era of advanced technology, olfaction remains a critical channel for person to conduct danger discrimination, emotional experience, and memory formation.
      • However, most substances in nature exist as multi-molecule mixtures.
    • Key English Points:
      • arXiv:2609.11958v1 Announce Type: new
      • Abstract: Olfaction played an indispensable role throughout human evolution and civilization
      • Even in the contemporary era of advanced technology, olfaction remains a critical channel for person to conduct danger discrimination, emotional experience, and…
      • However, most substances in nature exist as multi-molecule mixtures

  • Space as an Interventional Invariant: Cross-Modal Predictive Geometry for Stratified Cities and Em-Spaced Intelligence

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11959v1 Announce Type: new.
      • Abstract: Space is a foundational concept across mathematics, physics, spatial cognition, urban science, and embodied intelligence, yet these fields often treat spatial structure either as a shared geometric container or as a collection of disconnected representations.
      • Such approaches struggle to explain how heterogeneous sensory and urban processes can jointly reveal a common spatial structure, particularly when different modalities do not share the same metric or representation.

  • This paper addresses this gap by defining space as an interventional invariant: the minimal relational structure that preserves local compatibility and the conditional laws of future observations under admissible actions.

    • EN Highlights:
      • arXiv:2609.11959v1 Announce Type: new
      • Abstract: Space is a foundational concept across mathematics, physics, spatial cognition, urban science, and embodied intelligence, yet these fields often treat…
      • Such approaches struggle to explain how heterogeneous sensory and urban processes can jointly reveal a common spatial structure, particularly when different mod…
      • This paper addresses this gap by defining space as an interventional invariant: the minimal relational structure that preserves local compatibility and the cond…

  • On-Device Language Models for Privacy-Preserving Stress Prediction: A Multimodal Evaluation on Mobile Health

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Abstract: [To be translated] - arXiv:2609.11961v1 Announce Type: new.
      • Abstract: Stress is a pervasive determinant of mental health and a key target for mobile health interventions.
      • On-device language models (ODLMs) offer privacy-preserving inference without cloud dependency, yet their feasibility for health prediction under mobile resource constraints remains underexplored.
      • We evaluate ODLMs for multi-modal stress prediction using zero-shot prompting, measuring predictive accuracy alongside latency and throughput.
    • EN Highlights:
      • arXiv:2609.11961v1 Announce Type: new
      • Abstract: Stress is a pervasive determinant of mental health and a key target for mobile health interventions
      • On-device language models (ODLMs) offer privacy-preserving inference without cloud dependency, yet their feasibility for health prediction under mobile resource…
      • We evaluate ODLMs for multi-modal stress prediction using zero-shot prompting, measuring predictive accuracy alongside latency and throughput

  • FINESSE: An Agent-Based Simulator and Benchmark Dataset for Multimodal Financial Event Sequences

    • Publication Time: 2026-09-14 12:00 Beijing Time
    • Abstract: arXiv:2609.11993v1 Announce Type: new.
      • Abstract: Machine learning research in financial services is limited by the scarcity of representative open-source datasets.
      • Existing resources are often narrowly focused on a single modality or task and fail to reflect the structured, multimodal, and dynamic nature inherent to many problems in financial services.
      • In this paper, we introduce FINESSE, a Financial Event Sequence Simulation Environment, an agent-based simulation framework for generating synthetic, structured datasets composed of multiple interdependent event streams.
    • EN Key Points:
      • arXiv:2609.11993v1 Announce Type: new
      • Abstract: Machine learning research in financial services is limited by the scarcity of representative open-source datasets
      • Existing resources are often narrowly focused on a single modality or task and fail to reflect the structured, multimodal, and dynamic nature inherent to many p…
      • In this paper, we introduce FINESSE, a Financial Event Sequence Simulation Environment, an agent-based simulation framework for generating synthetic, structured…