2026-09-23

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

👍 154

09/21 08:00

An LLM agent's capability is largely magnified by its harness, namely the prompts, control flow, tooling, memory, and context management surrounding the frozen backbone model. Recent methods increasingly automate this process by iteratively proposing and selecting component-wise edits of an agent ha

中文介绍 RRSI 针对 LLM agent 的 harness（prompt、控制流、工具、记忆与上下文管理）自动改进问题：在冻结骨干模型下递归地提出并筛选组件级编辑，并以正则化约束抑制自我改进中的退化与漂移，使 agent 性能可持续迭代提升。

Transferring the Intelligence of VLMs to Robotic Control

👍 101

09/19 08:00

Humans can seamlessly adapt to both physical and digital worlds, suggesting that while a digital-to-real gap exists in embodiment, environment and task, human intelligence itself may transfer across this gap. This naturally raises a fundamental question: can the intelligence of vision-language model

中文介绍 论文探讨 VLM 的智能能否迁移到机器人控制：人类能无缝适应物理与数字世界，说明数字到现实虽存在具身、环境与任务差异，智能本身或可跨域迁移。研究据此把视觉语言模型的推理能力转移到机器人控制策略中。

OmniEdu: Open Foundation Models for Learning and Teaching

👍 70

09/19 08:00

Educational foundation models must solve problems, understand curriculum structure, diagnose learner difficulties, and provide appropriate instructional support. Existing educational language models often focus on either problem solving or tutoring, with training mixtures organized by source or task

中文介绍 OmniEdu 面向教育场景提出开放基础模型，要求同时完成解题、理解课程结构、诊断学习者困难并给出教学支持。针对现有教育模型偏重解题或辅导单一方向、训练数据按来源组织的问题，其数据按教学需求组织，兼顾学习与教学任务。

Document Retrieval-Aware Chunking (D-RAC): Universal Retrieval-Aware Ingestion of Enterprise Documents via PDF Normalization and Multimodal Markdown Conversion

👍 48

09/21 08:00

Retrieval-Augmented Generation (RAG) systems over enterprise knowledge bases must ingest heterogeneous document formats -- PDFs, Word documents, presentations, and scans -- whose content is locked inside complex visual layouts, multi-column pages, and dense tables. Rule-based extraction and OCR dest

中文介绍 D-RAC 解决企业文档 RAG 的摄取难题：PDF、Word、演示文稿与扫描件内容锁在复杂版面、多栏与密集表格中，规则提取与 OCR 易失真。方法先做 PDF 规范化与多模态 Markdown 转换，再按检索需求分块，提升知识库检索质量。

VideoGen-Agent: Reinforcing Video Generation Agents

👍 35

09/21 08:00

Recent advances in video generative models have enabled high-fidelity, temporally coherent video generation. However, these models often struggle to satisfy prompts requiring specialized knowledge, specific identities, physical consistency, or ordered events. In this paper, we present VideoGen-Agent

中文介绍 VideoGen-Agent 针对视频生成模型难以满足专业知识、特定身份、物理一致性与有序事件等提示的问题，提出以强化学习驱动的视频生成 agent 框架，通过 agent 化的规划与调用流程，提升复杂提示下的生成成功率与一致性。

OmniVChat: Synthesizing, Benchmarking, and Training for Native Audio-Visual Dialogue

👍 31

09/18 08:00

We define OmniVChat (Omni Video Chat) as the task of native audio-visual dialogue between a user and an omni model. In OmniVChat, omni models directly and simultaneously receive audio and video from a user and return text. The user's query is embedded in the audio and video, without a separate text

中文介绍 论文定义 OmniVChat 任务：omni 模型同时接收用户的音频与视频输入，查询直接嵌入音视频而无需单独文本，模型输出文本回答。工作覆盖数据合成、基准构建与模型训练，推动原生音视频对话这一新范式的发展。

onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction

👍 27

09/21 08:00

We present onPanda, an interactive tool for efficiently annotating LLM alignment data and agent trajectories. onPanda adopts token-level correction as its core interaction: while reading a model response, the annotator locates the first inappropriate token and either picks a substitute from the mode

中文介绍 onPanda 是面向 LLM 对齐数据与 agent 轨迹的高效标注工具，以 token 级纠正为核心交互：标注者阅读模型回复时定位第一个不当 token，并从候选中选取替代，从而以更低人工代价获得高质量偏好与纠正数据。

Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design

👍 25

09/18 08:00

Professional graphic design is a long-horizon agentic task in which structured, editable artifacts emerge from many interdependent actions, yet outcomes admit no reliable programmatic oracle. We introduce a continual adaptation framework in which a frozen frontier model operates professional design

中文介绍 Designer-RSI 面向专业平面设计这一长时程 agent 任务：产出的可编辑结构化稿件依赖大量相互关联操作，却缺乏可靠程序化判分。框架让冻结的前沿模型从用户流量中持续演化程序性记忆，实现持续适应与迭代改进。

Grounded Action Model: 3D Grounding as a Foundation for Robotics

👍 21

09/20 08:00

Manipulation policies must know which objects matter and where they are, yet the pretrained backbones that current robot foundation models build on, from language in vision-language-action models (VLAs) to video generation in world-action models (WAMs), do not directly require this metric grounding,

中文介绍 论文提出 Grounded Action Model，把 3D grounding 作为机器人操作的基础能力。现有 VLA 与 world-action model 的预训练骨干并不要求明确的空间与物体接地，GAM 以度量化的 3D 定位为前置，为策略提供「哪些物体重要、位于何处」的表征。

One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents

👍 16

09/20 08:00

Repository-level software engineering (SWE) comprises heterogeneous task categories, whose progress under pooled agentic reinforcement learning can be uneven: gains in some categories coincide with regressions in others, while aggregate resolution obscures these changes. Motivated by this category s

中文介绍 仓库级软件工程（SWE）任务类别异质，池化 agentic RL 下各类别进展不均甚至此消彼长，聚合指标掩盖了这种差异。论文提出类别感知的迭代专家训练方法，在「一对多、多对一」的专家组织下分类别提升 SWE agent 能力。

BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence

👍 15

09/16 08:00

Business intelligence (BI) is a cornerstone of enterprise decision-making and is widely used by enterprise users in software such as Power BI and Tableau. In traditional BI workflows, users need to prepare data by (1) identifying relevant tables, (2) performing data transformations, and (3) building

中文介绍 论文面向端到端商业智能自动化：传统 BI 流程需用户手工识别相关表、做数据转换并搭建可视化。工作提出 BI-Agent 与配套基准 BI-Bench，把表选择、数据变换到报表生成的完整链路交给 agent 自动完成并统一评测。

Realtime-Venus: A full-duplex interaction system with asynchronous delegation

👍 13

09/12 08:00

Natural interaction in digital and physical environments requires continuous perception and timely responses. Spoken dialogue relies on acoustic and linguistic cues, while video interaction also requires grounding the conversation in evolving visual context. We present Realtime-Venus, a proactive fu

中文介绍 Realtime-Venus 是全双工实时交互系统，通过异步委派机制实现持续感知与及时响应：语音对话依赖声学与语言线索，视频交互还需把对话锚定到不断变化的视觉场景中。系统支持主动式交互，兼顾低延迟与视觉语境理解。

Deep Persona: A Psychologically Grounded Architecture and Evaluation Framework for Role-Playing Agents and Simulations

👍 13

09/06 08:00

Existing approaches to persona simulation with Large Language Models (LLMs) mostly rely on shallow character descriptions that fail to sustain coherent character behavior across extended interactions. We introduce Deep Persona, a psychologically grounded, three-layered architecture that organizes pe

中文介绍 Deep Persona 指出 LLM 角色模拟多依赖浅层人设描述，难以在长程交互中保持行为一致。工作提出心理学基础的三层人格架构来组织角色属性，并配套评测框架，用于角色扮演 agent 与社会科学模拟中的一致性评估。

Harness-Zero: Harness Distillation via Agent-as-Harness

👍 12

09/21 08:00

Agent harnesses, the external systems that mediate model-environment interaction, can substantially improve agent performance, but their gains remain tied to the harness at deployment. Because the best harness varies across domains, instances, and models, a general-purpose agent must either settle f

中文介绍 Harness-Zero 关注 agent harness 的部署依赖：harness 能显著提升 agent 表现，但收益绑定在具体 harness 上，而最优 harness 随领域、实例与模型变化。方法通过 agent-as-harness 的蒸馏，把 harness 能力内化进模型本身。

Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents

👍 12

09/21 08:00

Agentic memory is becoming essential for long-horizon AI agents, yet many existing systems rely on autoregressive LLMs to control how memories are organized, retrieved, and used, placing expensive generation on the critical path of memory operations. We introduce \method, a new agentic memory archit

中文介绍 Jev-Mem 提出 System-One 控制的 agentic 记忆架构。现有系统用自回归 LLM 决定记忆的组织、检索与使用，把昂贵生成放在记忆操作关键路径上；该工作改由轻量机制控制记忆读写，降低延迟并提升长程任务效率。

CARE: Experience-Guided Atomic Corrective Execution for Vision-Language-Action Policies

👍 11

09/21 08:00

Vision-Language-Action (VLA) policies achieve strong performance in robotic manipulation but remain brittle once execution deviates from nominal trajectories. We propose CARE (Corrective Atomic Robotic Execution), a framework that improves recovery by learning from failures encountered during execut

中文介绍 CARE 针对 VLA 策略在机器人操作中一旦偏离标称轨迹即变得脆弱的问题，提出经验引导的原子纠正执行框架：从执行过程中遇到的失败中学习生成纠正动作，提升策略在扰动与偏差下的恢复能力。

1% of Tokens Can Be Enough: On Gradient Estimation in On-Policy Distillation

👍 9

09/21 08:00

Sparse on-policy distillation (OPD) allocates teacher supervision to a small subset of tokens in student-generated trajectories. However, useful teacher guidance can yield a noisy update when its gradient is estimated from a sampled next token. We study this estimation problem at a fixed prefix in i

中文介绍 论文研究稀疏 on-policy distillation（OPD）中的梯度估计问题：教师监督只施加于学生生成轨迹中的少量 token，用采样到的下一 token 估计梯度会引入噪声。在分析该估计偏差后提出改进方案，仅约 1% 的 token 即可提供有效监督。

ACLArena: Agent Continue Learning in Multi-stage Post-training

👍 8

09/21 08:00

Building general-purpose agents for industrial deployment requires integrating multiple capabilities, each typically acquired at a distinct stage of training. Yet there is currently no well-established recipe for Agent Continual Learning (ACL), with little understanding of the trade-offs among exist

中文介绍 论文提出 ACLArena，研究多阶段后训练下的 agent 持续学习（ACL）：工业级通用 agent 需在不同训练阶段依次获得多种能力，但各阶段间的权衡与相互干扰缺乏系统认识。该平台提供统一评测，厘清不同持续学习策略的取舍。

Gricea: An Open Science Platform for Conversational AI Research

👍 8

09/18 08:00

We need studies on conversational AI (CAI) at scale to understand human behavior and shape CAI design. However, fragmented reporting of systems and study configurations hinders replication, extension, and knowledge accumulation. We present Gricea, an open-science platform representing studies as con

中文介绍 Gricea 是面向会话 AI 研究的开放科学平台，把研究表示为可复用配置，解决系统与实验设置报告碎片化、难以复现与扩展、知识难以累积的问题，支撑大规模会话 AI 研究，以理解人类行为并指导系统设计。

MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup

👍 8

09/14 08:00

Scaling large language models efficiently has motivated sparse capacity mechanisms such as Mixture-of-Experts and, more recently, conditional memory: token-indexed embedding tables that augment the backbone with cheap parametric lookups. Existing memory-embedding methods retrieve via a deterministic

中文介绍 MoME 提出 Mixture-of-Memory Embeddings，用于上下文感知的稀疏查找：在 MoE 与条件记忆（token 索引嵌入表）基础上改掉确定性检索方式，让记忆检索依据上下文动态选择，以较低的参数开销提升模型容量与效果。

Salvor v1.0.0-beta: I got tired of my projects forgetting what they learned, so I taught the repo to

@BlockchainDan · 18.5K 粉丝 · 3.2M 阅 · 501 赞 · 35 转

09/21 13:25

Soft-launching an open-source, repo-native engineering knowledge layer for coding agents. Built because of one very bad /clear. About a year into building an HFT tool with AI agents, I typed one

中文介绍 开源项目 Salvor 发布 v1.0.0-beta：一个仓库原生的工程知识层，让 coding agent 记住项目里学到的经验。作者用 AI agent 写 HFT 工具约一年，因为一次糟糕的 /clear 导致上下文丢失而动手开发。

Jev is INSANE for Marketing

@dsqjaffa · 737 粉丝 · 217.8K 阅 · 510 赞 · 37 转

09/21 23:15

You can't escape it. Literally EVERYWHERE you look on your feed is filled with something about "jEv is iNsAnE"... and it's generated tens of millions of views on X in less than a week. I'm guilty of

中文介绍 拆解 Jev 在营销圈的刷屏现象：不到一周时间，时间线上到处都是「Jev 太疯了」的帖子，在 X 上累计数千万次浏览。作者从营销角度分析这波传播是怎么起来的，以及自己也没能免俗。

You're Already a Meat Proxy

@obie · 13.1K 粉丝 · 157.2K 阅 · 738 赞 · 98 转

09/21 05:02

I’m excited about a future that I suspect will be very hard on many long-time friends and certain people that I love. The same technology that is allowing me to be more successful than ever is

中文介绍 一篇偏思辨的短文：作者说自己对某个未来感到兴奋，但也清楚这个未来会让许多老朋友和所爱之人过得很艰难——同一个让他前所未有成功的技�术，正在把人变成「肉身代理」。不是教程，是对 AI 替代的感慨。

I reverse-engineered Instinct's memory. Here's exactly how it works

@DhravyaShah · 62.1K 粉丝 · 120.2K 阅 · 509 赞 · 16 转

09/21 02:49

Instinct has taken the world by storm over the last two weeks - it's one of the best imessage assistants I've used. as with every product, I decided to reverse-engineer Instinct's memory to find out

中文介绍 Instinct 近两周爆火，作者称它是自己用过最好的 iMessage 助手，于是逆向分析了它的记忆机制，把内部实现方式完整拆出来。属于产品逆向 + 记忆架构解读，适合想看 agent 记忆设计的人。

Jev + graphical models: a paradigm shift?

@fdellaert · 16.7K 粉丝 · 74.9K 阅 · 504 赞 · 61 转

09/21 08:19

Could Jev + graphical models fundamentally change how we build systems that reason under uncertainty? Unless you've been off the grid this week, you now know that Jev is @typesafeai’s new model for

中文介绍 作者提出一个开放问题：把 Jev（typesafeai 新模型）与概率图模型结合，会不会根本改变我们构建不确定性推理系统的方式？属于理论方向的探讨，不是教程或实测。

A Beginner's Guide to Jev

@omarsar0 · 320.6K 粉丝 · 73.5K 阅 · 523 赞 · 60 转

09/21 04:44

As you may have seen recently, Jev (a generalist System One model) took Twitter by storm. Many misleading examples are everywhere, but they don't really show Jev's real capabilities. Hence, we

中文介绍 Jev 刷屏后大量示例存在误导，作者写了一份入门指南，试图讲清这个通用 System One 模型真正的能力边界。定位是新手科普，纠正流传的错误演示。

Jev Explained for Normies

@matthewcanham · 10.7K 粉丝 · 57.1K 阅 · 516 赞 · 45 转

09/22 00:46

If you’ve been hanging out on the internet the past few days, you’ve probably come across Jev, a new AI model that’s taken the AI community by storm. Jev is not a traditional LLM. It’s built for

中文介绍 面向圈外人的 Jev 科普：这个模型几天内席卷 AI 社区，但它不是传统 LLM，而是为另一类任务设计的。作者用通俗语言解释它到底新在哪里、为什么值得关注。

Build your own Jev (100% local)

@_avichawla · 77.5K 粉丝 · 37.8K 阅 · 593 赞 · 80 转

09/20 14:46

Everything you need to turn an open-source LLM into a fast, local decision engine without retraining it. It covers next-token scoring, fixed choices with probability distributions, SGLang, and a

中文介绍 教程：不改模型权重，把开源 LLM 变成本地快速决策引擎的完整方案。涉及 next-token 打分、固定选项的概率分布、SGLang 等内容，核心是用概率分布取选项而非生成文本，全程可 100% 本地运行。

Build a Jev Judge

@akshay_pachaar · 289.5K 粉丝 · 32.1K 阅 · 501 赞 · 59 转

09/22 01:26

We use LLMs to write answers, then call another LLM to judge them. But when the judgment is a handful of bounded decisions, do we need another round of text generation? An agent answers a customer

中文介绍 教程思路：用 LLM 生成答案、再让另一个 LLM 打分是常规做法，但当判断只是几个有界的决策时，没必要再跑一轮文本生成。作者演示如何搭一个 Jev 式判断器，例如让 agent 处理客户问题分类。

A Swarm of Fruit Flies, Building a Civilization On-Chain

@murmur_arc · 983 粉丝 · 11.8K 阅 · 553 赞 · 137 转

09/20 21:17

Prologue — No LLMs, Only Neurons This project is called murmur, and it is alive right now on Arc mainnet (Chain ID 5042), settling in real USDC. To date the flies have completed more than 23,000

中文介绍 项目 murmur 已在 Arc 主网（Chain ID 5042）真实运行、以 USDC 结算。它不用 LLM，只靠神经元驱动的果蝇群体在链上自组织，至今已完成超过 23000 笔交易，属于链上自主 agent 实验。

🔬 An Oscar, Two Asteroids, and the Algorithm in Your sklearn: John Platt on AI for Science

Latent Space · 09/23 05:07

We talked to Google’s Oscar winning “Giganerd” about automating science, solving climate change, and how future generations can contribute to science in the age of superintelligent AI

中文介绍 Latent Space 对话谷歌奥斯卡获奖工程师 John Platt，讨论科学自动化、气候变化问题的解决，以及超级智能 AI 时代后代如何参与科学研究。

Better prompt caching for GPT-6

OpenAI News · 09/23 05:00

Learn how GPT-6 improves prompt caching with higher cache hit rates, new diagnostics, explicit breakpoints, and controls that reduce latency and costs.

中文介绍 OpenAI 介绍 GPT-6 的提示缓存改进：缓存命中率更高，新增诊断工具、显式断点及可降低延迟与成本的控制选项。

Introducing GPT-6 Sol and Luna

OpenAI News · 09/23 02:00

Meet GPT-6 Sol and Luna, two models that bring frontier intelligence to everyday work with different balances of capability and cost.

中文介绍 OpenAI 发布 GPT-6 Sol 与 Luna 两款模型，在能力与成本之间采取不同的平衡，将前沿智能引入日常工作场景。

Roundtables: The Deadly Failures of The Virtual Border Wall

MIT Tech Review AI · 09/22 21:42

The US has spent billions building a “virtual wall” of surveillance towers along its southern border over the past 25 years, promising they will help detect and apprehend border crossers and save lives. But a groundbreaking investigation by MIT Technology Review has documented over a thousand people

中文介绍 MIT 科技评论调查显示，美国过去 25 年斥资数十亿美元在南部边境建造由监控塔组成的「虚拟墙」，宣称可侦测并拦截越境者、挽救生命，但调查记录了其致命失败。

Parallel cut research time and cost in half with GPT‑6 Astra

OpenAI News · 09/22 20:00

GPT‑6 Astra allowed Parallel’s agents to research and synthesize labor-market data in half the time and at half the cost vs. prior models.

中文介绍 OpenAI 称，借助 GPT-6 Astra，Parallel 的智能体以此前模型一半的时间和一半的成本完成劳动力市场数据的研究与综合。

Don’t be fooled by this summer of AI hype

MIT Tech Review AI · 09/22 19:04

It’s been a busy few months for AI hype. At the end of April, Anthropic claimed that its model Claude Mythos is better at finding software vulnerabilities than most security experts. Then we had the OpenAI–Hugging Face hacking incident, after which Anthropic (proudly) and Meta (reluctantly) disclose

中文介绍 MIT 科技评论梳理今夏 AI 炒作：4 月底 Anthropic 宣称其模型 Claude Mythos 在发现软件漏洞方面强于多数安全专家，随后发生 OpenAI 与 Hugging Face 遭黑客攻击事件。

[AINews] Xiaomi MiMo-V2.6-Pro 1T-A42B: the new top Open Weights model, trained for $3M

Latent Space · 09/22 14:30

crowning a new Chinese frontier lab

中文介绍 小米发布 MiMo-V2.6-Pro 1T-A42B，成为新的顶级开放权重模型，训练成本约 300 万美元，被视为新的中国前沿实验室崛起。

Priorities and principles for effective third party assessments

OpenAI News · 09/22 08:00

OpenAI outlines priorities and principles for rigorous, secure, and independent third-party AI safety assessments of frontier models and safeguards.

中文介绍 OpenAI 提出对前沿模型与安全防护开展严格、安全且独立的第三方 AI 安全评估的优先事项与原则。

How UK AISI and EvalEval Are Making Benchmark Results Reproducible

Hugging Face Blog · 09/22 08:00

中文介绍 Hugging Face 博客介绍英国 AI 安全研究院（AISI）与 EvalEval 如何推动基准测试结果实现可复现。

Transformers now runs llama.cpp quants

Hugging Face Blog · 09/22 08:00

中文介绍 Hugging Face 博客介绍，Transformers 现已支持运行 llama.cpp 的量化格式。

Jun Kim, oMLX creator and maintainer, joins Hugging Face to support the MLX community

Hugging Face Blog · 09/22 08:00

中文介绍 oMLX 的创作者兼维护者 Jun Kim 加入 Hugging Face，以支持 MLX 社区。

Opus 5.5 imminent ⏳, Grok 4.7 🚀, MiMo v2.6 🤖

TLDR AI · 09/22 08:00

中文介绍 TLDR AI 汇总当日动态：Opus 5.5 或将发布、Grok 4.7 上线，以及小米 MiMo v2.6 等模型消息。

Jev: System One models for Prod, not God — with Diogo Almeida, CEO, TypeSafe AI

Latent Space · 09/22 06:13

The definitive Jev podcast with its lead creator.

中文介绍 Latent Space 对话 TypeSafe AI 首席执行官 Diogo Almeida，他亦是 Jev 的主要创作者，讨论面向生产环境而非「神级」的 System One 模型。

Pruning LLMs Like a Physicist: Block Removal as an Ising Optimization Problem

Hugging Face Blog · 09/21 21:44

中文介绍 Hugging Face 博客提出以物理学视角进行大语言模型剪枝，将模块移除问题转化为伊辛优化问题求解。

Higgsfield AI ships new video features in a day with GPT-6 Astra

OpenAI News · 09/21 20:00

With GPT-6 Astra, Higgsfield AI makes video ad creation easier for small businesses and brings new creative tools to market faster.

中文介绍 OpenAI 称 Higgsfield AI 借助 GPT-6 Astra 在一天内上线新的视频功能，让中小企业更易制作视频广告，并加快创意工具上市。