2026-10-05

RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations

👍 219

09/30 20:00

A companion that talks with a person for months should come to understand them. It should remember what they said, infer who they are, and know when the past bears on the message in front of it. Testing this requires a real person's record, and such records are private, so benchmarks generate the pe

中文介绍 论文提出了RealCompanion，一个基准测试，用于评估人工智能在长期对话中对人类理解的能力。研究通过使用真实对话记录，测试模型的记忆、推理和关联能力。这一基准具有重要的应用意义，尤其是在开发长期互动的对话代理方面，有助于推动人机交互的自然性与智能化。

Does Learning Protein Folding Generalize to Broader Reasoning?

👍 109

09/29 20:00

Large language models rely heavily on human text, which often conveys surface answers rather than the spatial and structural logic behind them. Protein folding is a natural testbed, because one solved structure yields thousands of exactly checkable spatial and topological statements. We ask: can lea

中文介绍 论文探讨了大型语言模型在蛋白质折叠任务中的泛化能力，提出了一种新的思路来利用表面答案与基础逻辑之间的关系。研究表明，模型在理解空间和结构方面的表现与其在蛋白质折叠方面的学习能力相关，为更广泛的推理能力提供了新的视角，影响生命科学领域模型的应用。

MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation

👍 85

09/28 20:00

Vision-language-action (VLA) models have advanced robotic manipulation, but their zero-shot generalization in new tasks and environments remains limited, and their reliance on specialized training keeps them from benefiting directly from rapidly advancing general-purpose vision-language models (VLMs

中文介绍 MotorMind提出了一种结合视觉、语言和动作的模型，以推动机器人在新任务与环境中的零样本操作能力。通过对一般视觉语言模型的利用，该方法显著提高了机器人的任务适应性，标志着机器人操作系统在多样化场景中的应用前景，尤其在智能家居与自动化领域。

On-Policy Parameter Update Direction Underlies Generalization in LLM Post-Training

👍 69

09/28 20:00

The strong generalization performance of on-policy post-training paradigms has motivated studies of their parameter update behaviors. However, these studies treat the observed behaviors only as byproducts in on-policy training, overlooking their potential to serve as optimization principles for impr

中文介绍 论文研究了在线策略后训练中参数更新行为的强泛化性能，提出了优化原则，以改进模型泛化能力。通过分析这些更新行为，研究为理解和提升大语言模型的训练方法提供了新的视角，对提高机器学习模型在复杂任务上的表现具有重要意义。

HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents

👍 43

10/01 20:00

We introduce HyperBrowseComp, a multilingual and multimodal browsing benchmark comprising 423 manually authored and human-validated questions across 13 languages, written by native or highly proficient speakers. Questions are designed to be extremely challenging. Each question targets a concise, pub

中文介绍 HyperBrowseComp是一个多语言、多模态的浏览基准，包含423个高难度人工编写的问题，旨在测试网页浏览代理的理解和处理能力。此基准在多样性和挑战性方面具有显著的优势，将推动网页交互系统在多文化环境中的适应性和智能化。

Pivot-SD: Efficient Self-Distillation for Masked Diffusion Language Models

👍 39

10/01 20:00

Masked diffusion language models (dLMs) offer a promising parallel alternative to autoregressive models for complex reasoning. However, they face a distinct credit-assignment challenge, since a few commitments during denoising sharply reduce the uncertainty over the remaining masked positions and sh

中文介绍 Pivot-SD提出了一种高效的自蒸馏方法，针对掩蔽扩散语言模型的信用分配挑战。通过优化模型在去噪阶段的行为，该方法在复杂推理任务中展现了优越的性能，推动了语言模型在推理与生成任务中的应用潜力，特别是在文本生成与理解领域。

PDE-JEPA: Predictive Representation Learning of Latent Dynamics Modeling for Parametric PDEs

👍 39

09/27 20:00

Physical trajectories contain more than snapshots of a system: they also reveal how its states evolve under governing conditions. However, representation learning for parametric partial differential equations (PDEs) has largely relied on reconstruction-based objectives that emphasize recovering obse

中文介绍 PDE-JEPA模型通过学习潜在动态实现参数偏微分方程（PDEs）的预测表示，强调物理轨迹和状态演变的关系。该研究为复杂物理系统的建模提供了新的思路，为物理模拟与工程应用带来潜在的影响，尤其在科学计算和动态系统分析中。

LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models

👍 36

09/29 20:00

Legal language models require reward signals that capture not only answer correctness but also the multidimensional quality of legal responses. Existing reward methods, however, often rely on coarse-grained holistic judgments, providing limited domain specificity and interpretability. We introduce L

中文介绍 LexReward提出了一种分类驱动的奖励框架，旨在提升法律语言模型的反馈质量。相较于传统评估方法，该框架提供了更细致的奖励信号，有助于改善法律文本生成的准确性和多样性，对法律科技领域的应用具有重要的推动作用。

Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers

👍 36

09/29 20:00

AI-generated content, often called AI slop, is increasingly common everywhere, particularly in academia. Slop in AI-generated scientific papers, however, has more complex patterns that cannot be easily detected by existing token-based AI detectors. Each part of such a paper looks plausible while the

中文介绍 论文针对AI生成科研论文中的内容质量问题，开展了对“AI混乱”的基准测试，分析其检测复杂性的原因。研究结果有助于推动学术领域对AI内容生成的规范与标准化，以提升科技交流的准确性与可靠性。

Latent-MOPD: Latent Multi-Teacher On-Policy Distillation

👍 35

09/30 20:00

On-policy distillation (OPD) trains a student on the responses it generates. Existing LLM multi-teacher OPD transfers what specialists predict through their output distributions. We introduce Latent-MOPD, to our knowledge the first representation-level multi-teacher OPD method for LLMs. It integrate

中文介绍 Latent-MOPD引入了一种新的多教师在线策略蒸馏方法，将代表学习与多教师输出传递相结合。此方法提高了大语言模型的学习效率，有助于在多样化任务中更有效地利用不同专家的知识，推动自主学习领域的发展。

SimuVerity: Benchmarking Agents for Engineering-Grade Simulink Model Generation

👍 34

09/30 20:00

Existing Simulink benchmarks mainly evaluate whether generated models compile, execute, or resemble a reference model. These criteria do not establish whether a model satisfies its engineering requirements. We introduce SimuVerity, a benchmark of 101 text-to-executable Simulink model-generation task

中文介绍 SimuVerity建立了一种新的基准，以评估生成的Simulink模型是否满足工程要求。该基准超越了传统的编译和执行评估标准，带来了对生成模型的深度解析，提升了工程领域中模型生成技术的可靠性与应用价值。

VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks

👍 33

09/30 20:00

As LLM agents undertake increasingly complex, long-horizon tasks, verifying their outputs becomes increasingly challenging. We study how verification capability can be strengthened with a fixed base model, without access to reference answers or grading rubrics at test time. Repeated sampling yields

中文介绍 VeriHarness研究了如何在不接触参考答案的情况下，利用固定基础模型验证长时间任务的输出。通过重复采样，该方法增强了验证代理的能力，为复杂任务中模型性能的提升提供了新的思路，对智能代理的应用具有显著影响。

Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory

👍 31

09/30 20:00

Long-video generation and world models have shown strong potential for interactive entertainment and embodied simulation by predicting future observations conditioned on user actions and historical memory. However, as memory sequences grow longer and their structures become increasingly complex, man

中文介绍 Spatial Memory Intelligence探讨了如何在生成长视频和构建世界模型中引入长时记忆，提升对用户互动的预测能力。论文提出了一种新策略，适应性强，适合于交互式娱乐和模拟领域应用，推动记忆驱动的模型发展。

Multilingual GSM-Symbolic: What determines capability transfer across languages?

👍 30

10/01 20:00

We understand little about how capabilities acquired in one language carry over to another, or what governs this transfer: evaluations rely on incomparable, saturation-prone datasets and rarely examine its determinants jointly. Identifying what predicts transfer would let us avoid exhaustive evaluat

中文介绍 Multilingual GSM-Symbolic研究语言能力转移的决定因素，提出了关键的预测模型，能够有效避免通用评价数据集的不足。该研究为跨语言能力提升提供了新的见解，预示着多语言处理技术的广阔应用前景。

Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems

👍 28

09/30 20:00

Scientific progress emerges from a longitudinal ecosystem in which researchers, institutions, funding agencies, collaboration networks, and the scientific literature co-evolve. As AI becomes increasingly involved throughout the scientific research cycle, understanding these interconnected and evolvi

中文介绍 论文讨论了学术研究生态系统中AI的影响，提出一个封闭回路的LLM模拟研究模型，以理解科研的演变与关联性。这一研究为提升科研效率和协作提供了新思路，具有重要的学术与实践价值。

ProAR: Learning Prospective Reasoning with Autoregressive Video Models

👍 19

10/01 20:00

Autoregressive (AR) video models excel at causal generation, but their reliance on next-chunk prediction confines them to a short-sighted, reactive paradigm. This limitation is particularly consequential for reasoning-oriented generation, where achieving a target outcome through valid intermediate s

中文介绍 ProAR提出了一种增强因果生成的自回归视频模型，旨在突破短期反应的限制，以支持目标导向的生成。该方法在视频生成与推理中的应用前景广泛，推动了智能系统在复杂任务中的执行能力。

Tail-Influence Sampling for CVaR Policy Evaluation

👍 17

09/28 20:00

Policies with similar mean returns can differ sharply in rare failures, yet estimating lower-tail conditional value-at-risk (CVaR) accurately can require many costly rollouts. When different conditional components of a stochastic workflow can be queried separately, we ask how to allocate a fixed eva

中文介绍 Tail-Influence Sampling研究了低尾条件风险价值（CVaR）策略评估中的抽样技术，提出了一种在固定评估条件下有效分配资源的方法。这一研究对于降低评估成本，提高决策准确性具有重要的应用价值，尤其在金融风险管理领域。

Local Support Learning

👍 17

09/30 20:00

We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of each weight matrix, we uncover a natural retention objective under which updates produced by gradient-based optimizers are suboptimal. Following this

中文介绍 Local Support Learning探讨了大规模预训练模型中的灾难性遗忘，提出了一种基于几何优化的保留目标，揭示了梯度优化中的不足之处。研究为改进模型的记忆力和持续学习提供了新视角，对持续学习领域的研究具有重要意义。

FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolution

👍 16

10/01 20:00

LLM-guided evolutionary methods, such as AlphaEvolve, have emerged as powerful approaches for challenging computational optimization problems, such as circle packing. However, prior work typically optimizes performance gain over a fixed number of iterations. We argue that practical optimization shou

中文介绍 FrugalEvo促进了成本感知的程序进化，通过优化迭代次数的性能，以平衡成本和性能，提高了程序演化方法的实用性。这一研究对于高效的计算优化问题解决方案具有重要的推动作用，尤其在算法工程领域。

Language Models that Play Chess and Explain Their Moves

👍 15

10/01 20:00

Modern chess engines are silent experts: they play at a superhuman level, but do not offer explanations for their play. On the other hand, language models (LMs) can generate plausible-sounding explanations, but their weak playing strength limits the utility of their explanations. We introduce Queen,

中文介绍 论文介绍了Queen模型，它使现代棋类引擎不仅能够进行高水平的下棋，还能解释其走法。通过结合语言模型与智能游戏技术，研究展示了人工智能在解释策略方面的潜力，为智能游戏和教育应用提供了新框架。

How To Hire A Grok Bot: Experience From The First Zero Human Company.

@BrianRoemmele · 489.9K 粉丝 · 1.1M 阅 · 533 赞 · 85 转

10/04 14:36

You do not build a Grok Bot. You hire one. The difference is the whole article. A chatbot answers a question and forgets the room. A Grok Bot has a name, a job, a conversation that persists, and a

中文介绍 博主讨论了如何雇佣 Grok Bot，这是一种具有人格化持续交流能力的聊天机器人，与传统的聊天机器人不同。博主强调了这一创新对企业运作的重要性，暗示实现这一技术的公司或将引领未来的商业模式。

He Who Controls the Compute Controls the Universe

@stevenmarkryan · 458.8K 粉丝 · 124.9K 阅 · 541 赞 · 84 转

10/05 01:25

Incredible things are happening in AI. What’s more incredible is that it’s still EXTREMELY early. As in, Day 0.000. Things are only going to get crazier from here. In case you’ve been living under a

中文介绍 博主指出人工智能领域正在经历重大变革，且这个时代仍处于初期阶段，未来的发展将会更加不可思议。他强调，当前是技术创新的关键时刻，预示着巨大的行业潜力。

Every language learner gets a tutor now

@TryLiveAvatar · 1.1K 粉丝 · 123.4K 阅 · 692 赞 · 188 转

10/05 14:45

The best way to learn a language was always a patient tutor. For the first time, everyone can have one. Here's what that changes for learners, teachers and schools. The hardest part is speaking Ask

中文介绍 博主分享了如何利用新技术为语言学习者提供个性化的辅导，提升学习效果。他指出，技术的介入将改变学习者、教师和学校的角色，为语言学习注入新的活力。

VIBE MANUFACTURING IS HERE

@gregisenberg · 724.2K 粉丝 · 44.1K 阅 · 569 赞 · 53 转

10/04 09:25

For the FIRST TIME in HISTORY, you can describe a physical product in one sentence and, a week later, hold it in your hands, cut from metal and built to your exact measurements. That's VIBE

中文介绍 博主介绍了一种新的产品制造方式，称之为 VIBE 制造，用户可以通过简单的描述在一周内获得定制的实体产品。这一概念象征着制造业的未来趋势，强调了个性化和快速响应市场的能力。

Evals: how to know whether an AI system actually works

@sermakarevich · 549 粉丝 · 33.0K 阅 · 520 赞 · 64 转

10/03 14:38

An article for everyone who ships, buys, or signs off on software built on large language models (LLMs): engineers, product managers, and the CEO. Written in plain language, from the big picture down

中文介绍 博主提供了关于如何评估大型语言模型（LLMs）AI 系统有效性的洞见，内容适合工程师和产品经理等各类从业者。强调了评估标准和方法的重要性，契合当下对 AI 产品质量的关注。

LIST OF HACKATHON FOR OCTOBER.

@onlyonealexia · 1.6K 粉丝 · 26.5K 阅 · 509 赞 · 53 转

10/03 22:08

A List of Web3, AI & Web2 Hackathons Compiled by @onlyonealexia There are a lot of hackathon lists that look impressive until you actually open the links. I compiled a list of ongoing and upcoming

中文介绍 博主整理了一份关于即将到来的 Web3、AI 和 Web2 的黑客马拉松活动清单，旨在为参与者提供机会。此举反映了社区对技术创新及合作开发的持续兴趣，助力新兴项目的发展。

All games are open source now

@rehan_shei · 11.9K 粉丝 · 24.9K 阅 · 520 赞 · 49 转

10/03 20:52

A week ago I pointed Opus 5.5 at Terraria before bed. I woke up to a fully working, reverse engineered version of the game. Then I added a homing missile launcher, a tactical nuke and a bunch of new

中文介绍 博主分享了利用 Opus 5.5 反向工程重新制作游戏的经历，展示了开放源代码的潜力。此内容突显了开源在游戏开发中的影响力，反映了行业对创作自由和社区协作的重视。

Reinforcement Learning from scratch #1

@goyal__pramod · 11.4K 粉丝 · 24.7K 阅 · 504 赞 · 59 转

10/03 00:45

Now I am aware you must have been looking forward to the second part of my CUDA blog (if you haven't read it yet, check it out!), but hey! A man can have varied interests. And I believe if you are

中文介绍 博主开启了关于从零开始学习强化学习系列的讨论，展现了对该主题的浓厚兴趣，尽管他同时还在进行其他技术内容的创作。这一分享体现出学习者对于新兴技术的渴求，也反映了AI领域普遍的学习热潮。

Connecting AI agents to enterprise knowledge

MIT Tech Review AI · 10/05 11:47

For all the data that AI systems continually amass and analyze, enterprise AI agents often suffer from a curious shortcoming: a lack of knowledge. More than data, knowledge is the understanding of what the data means in the context of individual organizations. AI agents need this understanding to re

中文介绍 企业AI代理面临缺乏知识的问题，即虽然能处理大量数据，但未能理解数据在特定组织中的意义。有效的AI代理需要具备这一知识，以充分发挥其潜力。

Our approach to EU text provenance rules

OpenAI News · 10/05 11:00

How OpenAI is approaching text watermarking under EU rules. Learn where watermarks apply, how detection works, and why access starts with researchers.

中文介绍 OpenAI在应对欧盟文本来源规则时，介绍了文本水印的应用、检测机制及研究人员如何获得访问权的相关信息。

Bringing predictive analytics to the agentic AI era

MIT Tech Review AI · 10/05 09:29

In 2026, the question for enterprise AI is no longer whether predictive models can outperform statistical forecasts—that argument is settled. The big question now is how to enable predictive systems to act on their own conclusions without drifting from business intent. The frontier has moved from pr

中文介绍 到2026年，企业AI的关键问题不再是预测模型是否优于统计预测，而是如何确保预测系统能够自主执行结论且符合商业目标。

Building advertising for the way people use AI

OpenAI News · 10/05 06:00

OpenAI introduces a new visual ad format in ChatGPT and expands measurement tools, attribution partnerships, and brand suitability for advertisers.

中文介绍 OpenAI推出新的ChatGPT视觉广告格式，并扩大了测量工具、归因合作及品牌适配性，以满足广告商的需求。

People really hate AI, so why can’t they get enough?

MIT Tech Review AI · 10/05 04:00

Over the summer I talked to the CEO of Springboards, a startup building an LLM that’s designed to come up with a wider variety of responses than its mainstream rivals do. At the start of the call, he said something that’s been stuck in my head since: “We often say that we’re a self-loathing AI…

中文介绍 一位初创公司CEO表示，尽管人们普遍对AI有抵触情绪，但这些技术的多样性让用户难以拒绝。

EmTech Future 2026: When AI Meets Everything

MIT Tech Review AI · 10/05 00:00

Yossi Matias, Vice President & Head of Google Research, explores how AI is beginning to reshape biology, infrastructure, manufacturing, and science, and why its greatest impact may come when it intersects with other fields. Step inside the newsroom with our MIT Technology Review editors for sharp an

中文介绍 谷歌研究副总裁Yossi Matias指出，AI正在重塑生物学、基础设施和制造业，并与其他领域交叉时可能产生重大影响。

Prime Inference 📈, agent population 📈, Claude academy 🎓

TLDR AI · 10/04 20:00

中文介绍 无人报道，今天是个平静的一天。

The Agent Said It Was Done. The Database Disagreed.

Hugging Face Blog · 10/03 18:56

中文介绍 无人报道，具体内容未披露。

[AINews] not much happened today

Latent Space · 10/03 04:45

a quiet day.

中文介绍 无人报道，今天发生的事情不多。

A model guide for the GPT-6 family

OpenAI News · 10/02 12:15

Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.

中文介绍 OpenAI提供关于如何选择和调整GPT-6模型、改进提示及技能、协调工具的指导，以便初创公司为生产做准备。

Redefining enterprise intelligence with autonomous AI

MIT Tech Review AI · 10/02 11:49

Enterprise AI is no longer a future ambition. It is in full operational flight. Model capabilities are advancing faster than most organizations can absorb, while the cost of performance continues to fall. Globally, AI investment is set to reach $2.5 trillion in 2026, up 44% from the previous year. F

中文介绍 企业AI已全面投入运营，其模型能力正在快速提升，投资预计到2026年将达到2.5万亿美元，较2023年增长44%。

Open-sourcing AstaBrief, the fast report-generation model in Asta

Hugging Face Blog · 10/02 11:19

中文介绍 无人报道，具体内容未披露。

Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience

Latent Space · 10/02 10:04

After leading Meta’s Llama models, Ahmad Al-Dahle is now transforming Airbnb with AI — from how its teams develop products to how it serves guests.

中文介绍 前Meta的Llama模型负责人Ahmad Al-Dahle正在通过AI转变Airbnb，从产品开发到客户服务的各个方面。

Don’t be fooled—LLMs don’t reason

MIT Tech Review AI · 10/02 04:00

On an afternoon in Seoul in March 2016, I watched a program I helped build put a stone on the fifth line of a Go board in what looked like a gift to its human opponent. Move 37 in game two of the five-game match looked so absurd that some commentators thought it was a…

中文介绍 该文探讨了大型语言模型（LLMs）在推理能力上的限制，通过实例分析其与Go游戏的关系。

[AINews] Pi 1.0, Pi Durable, and AIE NYC

Latent Space · 10/02 02:40

the minimalist harness goes stable... and TypeScript!

中文介绍 无人报道，具体内容未披露。