2026-07-02

Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models

👍 53

06/16 20:00

Embodied Vision-Language-Action (VLA) models are typically obtained by fine-tuning powerful pretrained VLMs on robotics data, yet it is unclear how much commonsense and factual knowledge they retain after adaptation. Failures on knowledge-sensitive tasks are ambiguous, conflating missing knowledge w

中文介绍 本文探讨了视觉-语言-行动（VLA）模型在从强大的预训练视觉-语言模型（VLM）微调到机器人数据后，如何保留常识和事实知识。通过在知识敏感任务上的失败分析，揭示了适应后知识的保持程度。这项研究有助于理解和改善VLA模型在实际应用中的表现，具有重要的实用意义。

Multi-Block Diffusion Language Models

👍 27

06/29 20:00

Block Diffusion Language Models (BD-LMs) improve diffusion-based text generation with KV caching and flexible-length generation. A natural next step is to extend them from Single-Block Diffusion (SingleBD) to Multi-Block Diffusion (MultiBD), where a running-set of consecutive blocks is decoded concu

中文介绍 提出了多块扩散语言模型（Multi-Block Diffusion Language Models），该方法通过键值缓存（KV caching）和灵活长度生成，提升了扩散文本生成的效果。本文从单块扩散模型扩展到多块，通过并行解码连续块来优化生成速度和质量，具有潜在的应用价值。

PerceptionRubrics: Calibrating Multimodal Evaluation to Human Perception

👍 26

06/25 20:00

We introduce PerceptionRubrics, a rubric-based evaluation framework that addresses the gap between saturated benchmark scores and real-world brittleness. Shifting evaluation from holistic semantic matching to rigorous atomic auditing, PerceptionRubrics pairs 1,038 information-dense images with over

中文介绍 引入PerceptionRubrics评估框架，以填补基准得分饱和与现实世界脆弱性之间的差距。通过与1,038个信息密集图像配对的严格审计方法，Shift了评估关注点，旨在提供更符合人类感知的多模态评估，推动了评估方法的发展。

SkillHone: A Harness for Continual Agent Skill Evolution Through Persistent Decision History

👍 23

06/22 20:00

Agent skills extend language-model agents with task-specific procedures, scripts, and references, but the tasks and environments they target continually change. Existing methods improve skills in bounded runs and retain only the final artifact, discarding the decision history that later agents need

中文介绍 针对不断变化的任务和环境，提出了SkillHone，这是一种持续智能体技能演变的框架。通过保留决策历史，该方法有助于后续智能体更好地适应新任务，提升应用于动态环境下的智能体表现，具有重要的应用前景。

TurboServe: Serving Streaming Video Generation Efficiently and Economically

👍 18

06/16 20:00

Streaming video generation is emerging as a new serving workload in which users interact with long-lived sessions that generate video progressively, chunk by chunk. Unlike offline video generation or typical LLM serving, streaming video generation must preserve session state across active and idle p

中文介绍 本文提出TurboServe，提升了流媒体视频生成的效率与经济性。该系统能够在用户交互的长时间会话中逐步生成视频，与传统视频生成方法相比，更能有效保持会话状态，具有广泛的应用潜力。

Managing Procedural Memory in LLM Agents: Control, Adaptation, and Evaluation

👍 18

06/21 20:00

Procedural memory is increasingly used to improve LLM agents on recurring workplace tasks, yet its ability to produce reusable skills remains poorly understood. We introduce AFTER, a benchmark of 382 realistic enterprise tasks spanning six professional roles and 22 procedural skills, designed to eva

中文介绍 引入AFTER基准，通过382个真实企业任务的评估，探索程序记忆在大语言模型（LLM）代理中的应用。研究结果表明，程序记忆对任务的重用性产生重要影响，为提高LLM在职场重复任务中的效能提供了新思路。

DataEvolver: Self-Evolving Multi-Agent Data Construction for Text-Rich Image Generation

👍 18

06/29 20:00

Text-rich image generation is one of the most challenging settings in image generation, since models must simultaneously produce visually realistic images and render legible, semantically aligned, and layout-consistent text. Existing data pipelines usually follow a static crawl-filter-freeze paradig

中文介绍 DataEvolver旨在针对文本丰富的图像生成进行自我进化的数据构建。通过产生视觉真实的图像及布局一致的文本，克服了静态数据管道的限制，推动了图像生成技术的发展，对多模态生成任务具有参考意义。

ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving

👍 17

06/30 20:00

In prefill-decode (PD) disaggregated LLM serving, each request is assigned to a decode worker after prefill. Existing decode routers balance only load; for mixture-of-experts (MoE) models this is incomplete: equally loaded workers can differ in latency, since each decode step loads the weights of ev

中文介绍 提出了ELDR，一种关注专家本地性的解码路由方法，用于弱化粒度的Mixture-of-Experts（MoE）模型服务。该方法改善了预填充-解码（PD）分解LLM服务的效率，具有加速处理请求的潜力。

MemSyco-Bench: Benchmarking Sycophancy in Agent Memory

👍 17

06/30 20:00

Memory has emerged as a cornerstone of modern LLM-based agents, supporting their evolution from single-turn assistants to long-term collaborators. However, memory is not always beneficial: retrieved memories often induce a critical issue of sycophancy, causing agents to over-align with the user at t

中文介绍 本文研究了记忆在现代LLM代理中的重要性，并提出了一种新的基准来评估记忆引发的拍马屁现象。此研究揭示了记忆整合的挑战，对如何平衡用户喜好与代理独立思考有助于深入了解，具有现实应用价值。

Are We Measuring Strategy or Phrasing? The Gap Between Surface- and Approach-Level Diversity in LLM Math Reasoning

👍 15

06/28 20:00

Diversity in LLM mathematical reasoning is critical for exploration, but common diversity metrics mostly capture surface-level variation rather than differences in how a problem is solved. We address this gap by introducing approach-level diversity: variation in strategies across correct solutions t

中文介绍 为了解决大语言模型（LLM）在数学推理中的多样性问题，本文提出了基于策略层面的多样性度量，超越了表面层面的变化。通过关注不同解决方案中的策略变异，推动了数学推理的探索性研究。

Multimodal Continuous Reasoning via Asymmetric Mutual Variational Learning

👍 14

06/30 20:00

Multimodal Large Language Models (MLLMs) are often constrained by a language-space bottleneck, forcing complex visual reasoning into discrete tokens which can lose perceptual nuance. A promising alternative is continuous latent reasoning, where the goal is to discover implicit reasoning pathways tha

中文介绍 提出多模态连续推理方法，旨在克服语言空间瓶颈。通过非对称互信息变分学习，探索隐含推理路径，实现复杂视觉推理的连续性，具有推动多模态大语言模型的潜力。

CausalMix: Data Mixture as Causal Inference for Language Model Training

👍 12

06/30 20:00

In Large Language Model (LLM) training, data mixing plays a pivotal role in determining model performance. Recent methods optimize mixture weights via proxy models, but they rely on the assumption of static data distributions. As a result, when the underlying data pool shifts, these methods require

中文介绍 CausalMix提出了一种将数据混合视为因果推断的方法，以改善大语言模型（LLM）训练。该方法针对动态数据分布，通过高效的混合权重优化，提升模型性能，为数据驱动的训练策略提供了新思路。

VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement

👍 11

06/30 20:00

As video corpora continue to expand in both scale and task complexity, there is increasing demand for approaches that retrieve relevant videos from large-scale corpora (inter-video reasoning) and subsequently perform fine-grained, query-conditioned tasks (intra-video reasoning) within the retrieved

中文介绍 VideoSearch-R1提出了一种迭代视频检索与推理的方法，通过软查询优化，解决了在大规模视频集合中检索相关视频和执行细粒度查询的需求，对视频检索技术具有重要的贡献。

Perceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning

👍 10

06/30 20:00

Fine-grained visual reasoning remains challenging for vision-language models, especially when small but critical visual cues are buried in high-resolution images. Existing approaches rely on repeated cropping or test-time visual search to introduce local evidence, but they typically do not explicitl

中文介绍 本文提出了Perceive-to-Reason框架，旨在解耦视觉感知与推理以增强细粒度视觉推理能力。通过对局部证据的引入，克服了现有方法在处理关键细节时的局限，推动了视觉语言模型的发展。

Little Brains, Big Feats: Exploring Compact Language Models

👍 10

06/28 20:00

While large language models have been dominating the research landscape recently, small language models remain highly relevant across various domains; yet, they receive far less attention. In this study, we investigate how smaller language models perform during the generation stage within a Retrieva

中文介绍 本文研究了小型语言模型在生成阶段的实际表现，强调其在多个领域的相关性，尽管相较于大型模型关注度较低，提出的研究为小型模型的应用与发展提供了更多视角。

Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity

👍 9

06/29 20:00

We present Seed2.0, a model series that takes a meaningful step toward solving complex, real-world tasks. Our approach begins with identifying users' genuine needs and constructing a reliable, forward-looking evaluation system by selecting and abstracting benchmarks grounded in these needs and in re

中文介绍 Seed2.0模型系列旨在为复杂现实任务提供解决方案，聚焦于识别用户真实需求并通过可靠的评估系统进行基准选择。这一方法有助于推动智能模型在实际应用中的可行性与可靠性。

ASPIRE: Agentic /Skills Discovery for Robotics

👍 9

06/29 20:00

Traditional robot programming is challenging: it requires orchestrating multimodal perception, managing physical contact dynamics, and handling diverse configurations and execution failures. We introduce ASPIRE (Agentic Skill Programming through Iterative Robot Exploration), a continual learning sys

中文介绍 ASPIRE提出了逐步机器人探索的代理技能编程框架，旨在解决传统机器人编程中的挑战。通过多模态感知和对动态环境的适应，提升机器人在复杂任务中的能力，具有潜在的实际应用价值。

QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents

👍 9

06/29 20:00

LLM agents increasingly act over long horizons, where a single trajectory can contain hundreds or thousands of actions. In these settings, outcome-only rewards provide too sparse guidance, failing to inform the model about the goodness of intermediate actions. Dense supervision methods aim to solve

中文介绍 QVal提出了评估长序列LLM代理的稠密监督信号方法，解决了在单一轨迹中获得稀疏反馈的问题。该研究对于动态任务中的有效监督尤为重要，推动了LLM在复杂环境中的应用研究。

MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training

👍 8

06/28 20:00

Modern large language models (LLMs) rely on reinforcement learning during post-training to push specific capabilities, yet integrating multiple capabilities into one model remains hard. Existing methods, such as Off-Policy Finetune and Mix-RL, are either inefficient or lose performance. In this work

中文介绍 MOPD通过多教师在线策略蒸馏的方法，解决了大语言模型（LLM）在后训练过程中集成多种能力的难题。此方法在提高效率和保持性能之间取得了平衡，对LLM的能力整合具有重要意义。

The State-Prediction Separation Hypothesis

👍 7

06/30 20:00

Transformers use the same forward computation stream to both predict the next token and store useful state for future token predictions. We formulate the state-prediction separation hypothesis: disentangling the two roles yields better language modeling performance. We design a Transformer variant t

中文介绍 本文提出了状态预测分离假设，认为将状态存储与下一个token预测解耦可以提升语言建模性能。该假设针对提升变换器模型的效果进行了相关实验，具有推动模型优化的潜力。

The AI Economy: The Next Chapter

@rickyho_1989 · 9.7K 粉丝 · 296.6K 阅 · 508 赞 · 69 转

06/30 08:23

Part I: The Economics of Intelligence Why the AI industry is about to optimize for intelligence per dollar rather than intelligence itself I have become increasingly convinced that the artificial

中文介绍 博主探讨了人工智能行业未来的经济模式，认为行业将更注重每美元的智能优化，而非单纯追求智能本身。这表明AI产业正在朝着经济效益和效率的方向发展。

How To Master Fable (Fundamentals Guide)

@milesdeutscher · 671.1K 粉丝 · 243.8K 阅 · 500 赞 · 73 转

07/01 14:13

TL;DR: Everything you need to do to get maximum value from Fable. I guarantee that after you're done reading, you'll have all the necessary tools to quite literally 10x your AI productivity with

中文介绍 这篇文章提供了使用Fable工具的完整指南，承诺能帮助用户将AI生产力提升十倍。内容涵盖了实现最大价值所需的所有基本元素，适合希望提高工作效率的用户。

ORACLE: Official AI Agents Trade on Polymarket

@OracleAiTrading · 34.1K 粉丝 · 176.1K 阅 · 2.7K 赞 · 567 转

06/29 14:06

In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We

中文介绍 博主分享了自主AI代理在2026年如何成为预测市场中的有效策略，目前超过30%的Polymarket活动来自算法和AI驱动的钱包，这反映了AI在金融预测领域的深化应用。

How To Become An AI Engineer in 2026 (Without a CS Degree)

@cyrilXBT · 187.0K 粉丝 · 171.8K 阅 · 505 赞 · 91 转

06/29 10:38

There is a sentence sitting on almost every AI engineering job posting that stops people before they even apply. Bachelor's degree in Computer Science required. Most people read that line, close the

中文介绍 新兴的AI工程岗位要求常常需要计算机科学学位，导致许多人在申请前就却步。博主分享了不具备该背景但依然可以成为AI工程师的途径，显示出行业对技能多样化的重视。

Wiki Memory

@hwchase17 · 115.7K 粉丝 · 129.7K 阅 · 500 赞 · 57 转

06/30 10:26

Memory for agents is still early, with little to no standards. “Memory” means something different to everyone. But one common pattern is emerging: wiki memory. The idea is simple: use an agent to turn

中文介绍 博主指出，虽然目前对AI代理的记忆功能仍处于初期阶段，但‘维基记忆’的概念逐渐崭露头角，意味着利用代理来优化信息管理和记忆。这是AI记忆领域一个新的发展趋势。

Toward Unrestricted Intelligence: Venice Series A

@ErikVoorhees · 908.5K 粉丝 · 124.8K 阅 · 529 赞 · 103 转

07/01 10:28

“If others possess your thoughts and constrain your words, then you exist at their permission, and you are not free.” —Anonymous Venice launched just over two years ago to create a private and

中文介绍 博主提到Venice项目创立的宗旨是为了实现思想的自由表达，探讨无约束智能的发展。这表明在AI快速发展中，自由与隐私问题正受到越来越多的关注。

46 thoughts on the near future

@bayeslord · 41.6K 粉丝 · 86.2K 阅 · 511 赞 · 43 转

06/30 16:37

This list is based on a thread I posted on June 4th. A few edits and additions here and there. Several people asked me to make the thread easier to read, so here it is. Intelligence I think people are

中文介绍 博主分享了关于未来的46个思考，基于其前期的讨论，这些观点旨在为读者提供对未来智能发展的更清晰的认识，反映出人们对AI影响的持续关注和思考。

What I learned helping set up 200 company brains

@jacob_posel · 28.7K 粉丝 · 54.5K 阅 · 501 赞 · 30 转

06/30 11:59

I have been directly or indirectly involved in helping set up over 200 company brains and context layers. Here are the three most important lessons. 1. The Biggest Predictor Of Success Is An Internal

中文介绍 博主分享了在帮助设立200个公司大脑中的经验，总结了成功的三个关键因素。这强调了在AI实现过程中内部合作的重要性，以及如何优化团队运作。

What Are Agents Paying For?

@base · 1.4M 粉丝 · 51.0K 阅 · 501 赞 · 112 转

07/01 02:43

Written by: @Must_be_Ash You've heard the slogans: "the agentic economy is here" and "agents are becoming the internet's newest paying customers." They sound big, but they skip the obvious question:

中文介绍 博主分析了‘代理经济’的兴起，提出了代理作为新兴消费者的重要性。这反映了AI代理角色转变带来的新机遇，引发广泛关注。

Building Infra for the Agent-Native Internet: June Recap, July Vision

@OptimaiNetwork · 99.8K 粉丝 · 7.8K 阅 · 540 赞 · 351 转

07/01 11:05

The AI industry is entering a new phase. The conversation is no longer centered on models alone. Increasingly, it is about infrastructure: how intelligent systems access knowledge, coordinate with one

中文介绍 AI行业正在进入新阶段，讨论逐渐从单纯的模型构建转向基础设施建设的关键性，这显示出如何让智能系统更高效地获取知识和资源的趋势。

Teaching AI to run with the turbines

MIT Tech Review AI · 07/02 08:51

Artificial intelligence may have captured the public imagination through chatbots and image generators, but some of its most consequential use cases are unfolding far from consumer-facing tools. In industries where physical infrastructure, operational continuity, and safety are paramount, AI is beco

中文介绍 人工智能在当前除了聊天机器人和图像生成器等消费工具外，还在工业领域展现出重要用途，涉及到基础设施、运营连续性和安全等方面。

[AINews] not much happened today

Latent Space · 07/02 03:10

another quiet day.

中文介绍 今天相对安静，没发生太多新闻。

AIEWF Daily Dispatch: Autoresearch and the tension between AI and human agency

Latent Space · 07/02 02:13

The software factory vision met resistance today from speakers defending human understanding and control.

中文介绍 软件工厂的愿景遭到反对，发言者强调人类理解和控制的重要性。

Gemini Flash upgrade ⚡️, Meta AI cloud 🌐, ZCode 👨‍💻

TLDR AI · 07/01 20:00

中文介绍 更新内容包括Gemini Flash和Meta AI云，但具体细节未提供。

Autoresearch: The feedback loop behind self-improving agents

Latent Space · 07/01 19:52

Introspection co-founder Roland Gavrilescu explains autoresearch, agent “recipes,” self-improving loops, and why humans remain central to the software factory.

中文介绍 自我研究的共同创始人Roland Gavrilescu讲解自我改进的循环及其重要性，强调人类在软件工厂中的核心地位。

How Cursor deploys AI inside the enterprise

Latent Space · 07/01 15:03

Cursor's Pauline Brunet explains how her team of Forward Deployed Engineers help organizations implement agents — essentially setting up software factories.

中文介绍 Cursor的Pauline Brunet解释了外派工程师团队如何帮助企业实施智能代理，设立软件工厂。

🔬 The Coolest Diffusion Research Isn't in LLMs — Evan Feinberg & Sergey Edunov, Genesis Molecular AI

Latent Space · 07/01 10:42

Why the Llama lead left Meta for drug discovery, PEARL's zero-shot OpenBind win, and what becomes possible when co-folding finally crosses the accuracy threshold.

中文介绍 讨论了Llama项目负责人为何离开Meta进入药物发现，以及PEARL在OpenBind零样本中的获胜。

LLMs are stuck in a groupthink groove. This startup is trying to get them out.

MIT Tech Review AI · 07/01 10:35

Let’s start with a game. Open up your chatbot of choice—Claude, ChatGPT, Gemini—and type “Give me a random number between 1 and 10.” You’re going to get 7. Almost always. Now type “Another” and you’ll get 3 or 4. Type “Another” again and you’ll get 8 or 9. That won’t work every time—but if it…

中文介绍 在测试下一步生成的随机数字时，发现大型语言模型（LLM）存在群体思维的问题，某创业公司正试图打破这一困境。

Warp CEO Zach Lloyd on why software factories are the next phase of coding

Latent Space · 07/01 10:28

Warp's founder thinks every major software project will soon run on an automated factory. He discusses why and how engineers should prepare for this shift.

中文介绍 Warp创始人认为未来每个主要软件项目都将依赖自动化工厂，他讨论了工程师如何为这一转变做准备。

not much happened today

Smol AI News · 07/01 01:44

**Anthropic** re-enabled **Claude Fable 5** with updated cybersecurity safeguards routing some requests to **Opus 4.8**. The relaunch influenced tooling adoption by **Cursor**, **Devin**, and **Perplexity**. Builders are adapting to frontier-model constraints by employing **multi-model orchestration

中文介绍 Anthropic重新启用了Claude Fable 5，并通过更新的网络安全措施影响了Cursor等工具的采纳。

AIEWF Daily Dispatch: Loops, Software Factories & Forward Deployed Engineers

Latent Space · 07/01 00:46

On Tuesday at the AI Engineer World's Fair, there was a lot of talk about loops, agent engineering, and the emergence of software factories. Also a hot topic: open models.

中文介绍 在人工智能工程师博览会上，讨论了循环、代理工程和软件工厂的出现等话题。

[AINews] Sonnet 5 today, and Fable 5 tomorrow

Latent Space · 06/30 23:01

Everything is open again!

中文介绍 今天再次开放Sonnet 5，明天将推出Fable 5。

Forward Deployed Engineers and the future of software engineering

Latent Space · 06/30 20:20

Sierra's Natalie Meurer on why product engineers and forward deployed engineers are starting to converge.

中文介绍 Sierra的Natalie Meurer探讨了产品工程师与外派工程师之间的融合现象。

Hugging Face and Cerebras bring Gemma 4 to real-time voice AI

Hugging Face Blog · 06/30 20:00

中文介绍 Hugging Face与Cerebras合作推出实时语音AI产品Gemma 4，具体细节未提供。

Claude Sonnet 5 🎭, Fable approved 🚀, Nano Banana 2 Lite 🍌

TLDR AI · 06/30 20:00

中文介绍 更新内容包括Claude Sonnet 5和Fable获批的信息，但具体细节未提供。