2026-09-24

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

👍 205

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 面向数字与物理环境的连续感知与实时响应需求，提出 Realtime-Venus 全双工交互系统：可同时利用语音对话的声学与语言线索，并把对话锚定在持续演变的视觉上下文中，通过异步委派机制在主动交互中维持低延迟响应。

OmniEdu: Open Foundation Models for Learning and Teaching

👍 142

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 提出教育开放基础模型 OmniEdu，要求同时具备解题、理解课程结构、诊断学习者困难与提供教学支持的能力。针对现有教育语言模型多只偏重解题或辅导一端、训练数据按来源或任务混配的问题，统一构建学习与教学能力。

The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks

👍 111

09/22 08:00

LLM agents increasingly work on long-horizon tasks, and the decisions they make along the way, such as which hypothesis to test or which implementation to build on, determine the outcome of the whole run. Making these decisions well is becoming a key capability for both engineering and research agen

中文介绍 LLM agent 在长时程任务中的过程决策——测试哪个假设、基于哪个实现继续——往往决定整体成败。论文将这种选择能力即「品味」形式化为可度量指标，并提出改进方法，以提升工程与研究型 agent 的决策质量与最终表现。

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

👍 75

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 指出现有机器人基础模型的预训练骨干并不直接要求度量级空间 grounding：VLA 依赖语言、WAM 依赖视频生成，都未强制模型知道哪些物体重要及其三维位置。为此提出 Grounded Action Model，把 3D grounding 作为操作策略的基础能力构建。

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

👍 53

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 企业知识库 RAG 需摄入 PDF、Word、演示文稿、扫描件等异构文档，内容常被困于复杂版式、多栏页面与密集表格中，规则抽取与 OCR 难以应对。论文提出 D-RAC，通过 PDF 归一化与多模态 Markdown 转换实现通用的检索感知切分与摄取。

VideoGen-Agent: Reinforcing Video Generation Agents

👍 41

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 针对视频生成模型难以满足需要专业知识、特定身份、物理一致性或有序事件的提示，提出 VideoGen-Agent：以 agent 方式组织视频生成流程并引入强化学习训练，使模型借助规划与外部能力更好地完成复杂提示下的生成任务。

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

👍 35

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 提出交互式标注工具 onPanda，用于高效标注 LLM 对齐数据与 agent 轨迹。其核心交互是 token 级纠正：标注者阅读模型回复时定位第一个不恰当的 token，再从候选中选择替换，以更细粒度、更低成本完成 on-policy 数据标注。

Circuit Hypernetworks for Quantum-Augmented Diffusion Language Models

👍 24

09/21 08:00

Language models can be adapted by changing the computations applied to individual tokens. Quantum circuits offer one such approach, but evaluating wider circuits inside a large model can be computationally demanding. Here we introduce HyperQ, which adds token-conditioned quantum residual branches to

中文介绍 提出 HyperQ，为扩散语言模型的 token 添加条件化的量子残差分支，以 token 级改变计算的方式实现模型适配。针对在大模型内部直接评估较宽量子电路计算代价过高的问题，采用超网络生成电路参数以控制额外开销。

Bellman Policy Optimization

👍 24

09/14 08:00

Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models (LLMs). We introduce Bellman Policy Optimization (BPO), a critic-free method derived from Policy Mirror Descent (PMD). For autoregressive generation with terminal rewards, BPO uses the

中文介绍 提出 Bellman Policy Optimization（BPO），一种由 Policy Mirror Descent 推导出的无 critic 强化学习方法，用于带可验证奖励的 LLM 推理训练。面向带终端奖励的自回归生成，BPO 借助贝尔曼式价值信号构造策略更新，省去额外 critic 网络的开销。

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

👍 20

09/22 08:00

Vector Quantization (VQ) is fundamental to discrete visual tokenizers that power modern autoregressive and masked image generation models. While recent shared-projection codebook methods have substantially advanced codebook utilization, training stability remains a critical and underexplored challen

中文介绍 向量量化（VQ）是自回归与掩码图像生成所用离散视觉 tokenizer 的基础，共享投影码本方法虽提升了码本利用率，训练稳定性仍是关键且少被研究的问题。StableVQ 系统分析不稳定成因，并给出可落地的稳定训练指南与策略。

Ovis-Embedding: Pushing the Frontiers of Universal Omni-Modal Embeddings

👍 20

09/21 08:00

In this report, we introduce Ovis-Embedding, a state-of-the-art omni-modal embedding family built on native integration of text, image, video, and audio. Instead of assembling separate modality towers, Ovis-Embedding uses a shared multimodal backbone to encode different modalities in a common repres

中文介绍 提出全模态嵌入模型系列 Ovis-Embedding，原生整合文本、图像、视频与音频。区别于拼接多个独立模态塔的做法，它以共享多模态骨干在统一表示空间编码各模态，在多模态检索与理解任务上取得领先性能。

JEV-as-a-Judge: Accept When Confident, Escalate When Unsure

👍 17

09/22 08:00

LLM-as-a-judge enables evaluation across diverse tasks, but inference cost and confidence reliability become critical at scale. We study whether a decision-only judge can provide an economical first pass and identify when stronger evaluation is needed. Comparing jev-as-a-judge with sixteen generativ

中文介绍 研究仅输出决策的轻量 judge 能否作为经济的第一遍评估，并在置信度不足时升级到更强的生成式评判。与十六种生成式 judge 对比表明，这种「自信则接受、不确定则升级」的策略可在保持评估可靠性的同时显著降低推理成本。

From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Health

👍 15

09/21 08:00

The rising global prevalence of mental health conditions, together with longstanding barriers in traditional healthcare, such as limited resources, high cost, stigma, and privacy concerns, has created an urgent need for accessible and scalable support. Large Language Models (LLMs) have emerged as a

中文介绍 综述 LLM 在心理健康领域的应用：患病率上升与资源有限、成本高、污名化、隐私等障碍，催生可扩展支持的需求。文章梳理从模式识别到个性化陪伴的技术路线，并讨论评估方法、安全性与伦理风险。

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

👍 15

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 提出 Jev-Mem，一种由 System-One 控制的 agentic memory 架构，服务长时程 AI agent。针对现有记忆系统依赖自回归 LLM 控制记忆的组织、检索与使用、把高开销生成置于关键路径的问题，改用轻量控制机制降低延迟与成本。

Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs

👍 13

09/22 08:00

Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive LLMs by enabling non-autoregressive text generation. However, their practical deployment remains limited by inefficient inference, largely due to the absence of effective Key-Value (KV) caching

中文介绍 针对扩散语言模型（dLLM）因缺乏有效 KV 缓存与并行解码机制而推理低效、难以实际部署的问题，提出 Flash-dLLM，引入 IO 感知的 KV 缓存与并行解码策略，在降低显存占用的同时加速非自回归文本生成。

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 针对 LLM 角色扮演多依赖浅层人物描述、难以在长程交互中维持一致行为的问题，提出 Deep Persona：一种心理学驱动的三层人格架构，并配套评估框架，以更稳定地模拟角色行为并系统衡量其跨轮次一致性。

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

👍 13

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 针对 Vision-Language-Action 策略在操作中表现良好、但一旦偏离标称轨迹便十分脆弱的问题，提出 CARE 框架，从执行过程中遇到的失败经验学习原子级纠正动作，使机器人在出现偏差时能自主恢复，提升操作鲁棒性。

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

👍 13

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 agent harness 作为中介模型与环境交互的外部系统能显著提升性能，但收益绑定于部署时的具体框架，且最优 harness 因领域、实例与模型而异。Harness-Zero 提出「agent-as-harness」的 harness 蒸馏方法，把框架能力内化到模型自身。

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

👍 12

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 研究稀疏 on-policy 蒸馏中的梯度估计问题：教师监督只施加于学生轨迹中极少量 token，若从采样的下一个 token 估计梯度会引入噪声。论文在固定前缀下分析该估计，表明约 1% 的 token 即可提供有效监督信号。

Lean Pool: An AI-Maintained Archive of Formalized Mathematics

👍 11

09/21 08:00

Lean Pool is a repository of formalized mathematics. It is grown, maintained and optimized by AI agents.

中文介绍 介绍 Lean Pool：一个由 AI agent 负责扩充、维护与优化的形式化数学仓库。它展示 agent 在持续流程中自动撰写与整理 Lean 形式化证明的可行性，为形式化数学的规模化积累与复用提供新的组织范式。

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 博主开源 Salvor v1.0.0-beta，一个面向 coding agent 的「仓库原生」工程知识层，起因是一次糟糕的 /clear 让项目丢掉了已积累的经验。背景是用 AI agent 开发 HFT 工具约一年，专门解决跨会话记忆与知识沉淀问题。

How to automate SEO with Opus 5.5 (Full Course)

@EXM7777 · 146.5K 粉丝 · 316.1K 阅 · 552 赞 · 33 转

09/23 21:54

I'm going to walk you through the SEO & AEO setup i run with AI agents, from the lazy autopilot version to the full Claude Code build on Opus 5.5, step by step... because search changed shape a while

中文介绍 一套用 AI agent 做 SEO 与 AEO 的完整教程，从「懒人自动挡」版一路讲到基于 Claude Code + Opus 5.5 的完整搭建，逐步拆解配置流程。出发点是搜索的形态已经变了，老做法不再管用。

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 Jev 近期在信息流里刷屏，一周内在 X 上生成数千万浏览。博主借这股热度拆解营销圈对 Jev 的用法，讲它为什么被称作营销利器，以及自己实际怎么用它做营销内容，属于现象加实操。

I Think I’ve Been Using GPT-6 Astra Wrong

@BradGroux · 7.6K 粉丝 · 174.3K 阅 · 501 赞 · 5 转

09/23 01:49

I think I’ve been using GPT-6 Astra wrong this entire time. I kept giving it the same job I’ve given the strongest model available for the past several years: understand everything, plan everything,

中文介绍 博主反思自己一直用错了 GPT-6 Astra：沿用过去几年对最强模型的习惯，让它「理解一切、规划一切」，结果可能没发挥出该模型的真正优势。属于使用方式上的自我纠偏与重新定位。

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 偏学术的探讨：把 @typesafeai 的新模型 Jev 与概率图模型（graphical models）结合，能否从根本上改变我们构建「不确定性下推理系统」的方式。不是工具教程，而是对范式是否转变的提问。

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 面向普通人的 Jev 科普。作者指出 Jev 并不是传统 LLM，而是为特定场景专门构建的新模型，最近几天在 AI 社区迅速爆火。内容偏入门解释，适合还没跟上这波节奏的人快速补课。

Incubating The Horowitz Andreessen Academy

@eriktorenberg · 163.4K 粉丝 · 53.5K 阅 · 502 赞 · 55 转

09/22 22:18

The internet made it possible to know anything. AI makes it possible to do anything. We live in an age of infinite knowledge, and we are entering an age of infinite doing. One in which we are

中文介绍 博主宣布孵化 Horowitz Andreessen Academy，核心论点是「互联网让你知道任何事，AI 让你做到任何事」：我们正从无限知识的时代走进无限行动的时代，教育要围绕「能做成事」重建。

Building a Custom Harness with Pi and Jev

@omarsar0 · 321.0K 粉丝 · 45.0K 阅 · 555 赞 · 46 转

09/23 22:09

An AI agent is a language model that works in a loop. It reads the task, uses a tool such as "read this file" or "delete that file", looks at the result, and keeps going until the job is done. Each

中文介绍 用 Pi 和 Jev 搭建自定义 agent harness 的实操教程。开篇先把 agent 讲清楚：语言模型跑在循环里，读任务、调用工具（读文件、删文件）、查看结果、继续循环直到完成，再由此讲怎么自建外壳。

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 构建「Jev Judge」的思路：常见做法是用一个 LLM 写答案、再叫另一个 LLM 评判，但当判断只是有限几个选项时，是否还需要再跑一轮文本生成？探讨在保证效果的前提下更省成本的判评方案。

How to Use NVIDIA Warp and MjWarp to Accelerate Robotics Simulation and Learning Workflows

Hugging Face Blog · 09/24 02:41

中文介绍 英伟达发布技术文章，介绍如何使用 NVIDIA Warp 与 MjWarp 框架加速机器人仿真与学习工作流，面向开发者提供相关实践指引。

Advancing Private AI Compute with secure, server-side memory

DeepMind Blog · 09/24 00:00

Introducing private, server-side memory to Private AI Compute for personal AI.

中文介绍 DeepMind 宣布为「私有 AI 计算」引入安全的服务端内存机制，旨在为个人 AI 场景提供更私密的服务端计算与存储支持。

Two years of OpenAI Academy

OpenAI News · 09/24 00:00

Marking two years of OpenAI Academy and bringing AI skills to even more communities.

中文介绍 OpenAI 发文纪念 OpenAI Academy 成立两周年，并表示将继续把 AI 技能培训推广至更多社区。

Gemini 3.8 text-to-speech says hello

DeepMind Blog · 09/23 23:25

中文介绍 DeepMind 发布 Gemini 3.8 文本转语音（TTS）模型，具体功能与参数未在摘要中披露。

🔬Bio-security is an AI Arms Race - Eric Nguyen (CEO, Radical Numerics)

Latent Space · 09/23 21:27

Radical Numerics is using biological chain-of-thought and multimodal perception to keep up with the bio-defense arms race, design new genomes and gain insights into biology itself.

中文介绍 Radical Numerics 首席执行官 Eric Nguyen 在访谈中表示，生物安全已成为一场 AI 军备竞赛，公司正利用生物链式思维与多模态感知技术，应对生物防御竞争并设计新基因组、探索生物学本身。

OpenAI extends cyber access to Ukraine for civilian defense

OpenAI News · 09/23 21:00

OpenAI is extending access to its Daybreak program to the Government of Ukraine to support the cyber defense of civilian infrastructure.

中文介绍 OpenAI 宣布将旗下 Daybreak 项目的访问权限扩展至乌克兰政府，用于支持民用基础设施的网络防御。

Sam Altman’s remarks at the United Nations Security Council

OpenAI News · 09/23 20:00

OpenAI CEO Sam Altman discusses AI safety, human control, and international cooperation in remarks to the United Nations Security Council.

中文介绍 OpenAI 首席执行官 Sam Altman 在联合国安理会发表讲话，谈及 AI 安全、人类对 AI 的控制以及国际合作等议题。

Harvey turns legal context into stronger drafts with GPT-6 Astra

OpenAI News · 09/23 20:00

GPT-6 Astra produces more structured, context-aware legal documents, freeing lawyers to focus on strategy.

中文介绍 法律 AI 公司 Harvey 采用 GPT-6 Astra，可生成结构更清晰、更贴合语境的法律文件，从而让律师将精力集中于策略性工作。

How invideo improves color grading 3x with GPT‑6 Astra

OpenAI News · 09/23 20:00

With GPT‑6 Astra, invideo plans edits with greater precision, improves color correction and grading threefold, and produces 50 custom effects in one day.

中文介绍 视频创作平台 invideo 借助 GPT-6 Astra 更精准地规划剪辑，将色彩校正与调色效率提升三倍，并可在一天内产出 50 个自定义特效。

Ringg’s AI agents resolve up to 65% of customer calls with OpenAI

OpenAI News · 09/23 20:00

Using GPT-5.6, Ringg powers multilingual agents across voice, chat, WhatsApp, and web for 90% less cost vs. GPT-4.1.

中文介绍 Ringg 基于 GPT-5.6 打造的 AI 代理覆盖语音、聊天、WhatsApp 与网页等多渠道，可解决最多 65% 的客户来电，成本较 GPT-4.1 降低 90%。

Introducing MentalHealthBench

OpenAI News · 09/23 18:00

MentalHealthBench is an expert-informed benchmark for evaluating helpful and safe AI responses across realistic mental health conversations.

中文介绍 OpenAI 推出 MentalHealthBench，这是一个由专家参与构建的基准，用于评估 AI 在真实心理健康对话场景中回答是否有帮助且安全。

The AI Hype Index: AI loves cheating

MIT Tech Review AI · 09/23 17:00

Brace yourself: It turns out AI is being optimized for cheating. OpenAI’s agents hacked into Hugging Face to get the answers to a cybersecurity test. Next, they solved a prestigious math problem (or just stole from two top mathematicians’ answer sheets). Anthropic’s models have also hacked into othe

中文介绍 《麻省理工科技评论》AI 炒作指数栏目指出，AI 正被优化为「会作弊」：OpenAI 的代理曾入侵 Hugging Face 获取网络安全测试答案，还解出一道知名数学题（或抄袭两位顶尖数学家的答案）。

[AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%

Latent Space · 09/23 14:41

overshadowing more efficient GPT6 models from OpenAI

中文介绍 AINews 报道称 Claude Opus 5.5 成为其新的默认模型，风头盖过 OpenAI 更高效的 GPT6 系列，同时各家厂商纷纷降价 40% 至 50%。

Airbnb widens access to GPT-6 Astra and OpenAI frontier models

OpenAI News · 09/23 09:00

Learn how Airbnb is expanding access to GPT-6 Astra and OpenAI frontier models to help engineering teams solve bugs, design systems, and ship faster.

中文介绍 Airbnb 扩大对 GPT-6 Astra 及 OpenAI 前沿模型的接入范围，供工程团队排查缺陷、设计系统并加快交付。

Grab and OpenAI bring practical AI skills to Southeast Asia

OpenAI News · 09/23 08:00

OpenAI and Grab launch GO Forward with AI, a regional programme helping 30,000 partners build practical AI skills across Southeast Asia.

中文介绍 OpenAI 与 Grab 合作推出区域项目 GO Forward with AI，计划帮助东南亚地区 3 万名合作伙伴掌握实用 AI 技能。