行业信号一手源NEW
1 小时前
The rules will come into force on Monday along with other changes to planning policy.
可信度 88交叉信源 1
话题
政策 / 监管
共 307 条相关资讯 · 来自历史归档
Changing pubs into homes to be made harder under new rules
行业信号
8/14 14:34
IT之家 8 月 14 日消息，SpaceX 完成了对 AI 编程初创公司 Cursor 的 600 亿美元 （IT之家注：现汇率约合 4,054.22 亿元人民币） 收购。 对于马斯克而言，这笔交易是追赶 Anthropic 和 OpenAI 的重要一步。监管文件显示，收购于当地时间 8 月 14 日正式生效。SpaceX 两个月前已经宣布与 Cursor…
行业信号
8/14 10:15
Human-AI marriages are not currently recognized by US law. Some Republican state policymakers are drafting legislation to keep it that way.
可信度 74交叉信源 1
行业信号
8/14 06:18
Skill / 资源
8/12 21:00
Open models may soon be added to an updated AI framework, sources tell WIRED, as the White House continues to grapple with how to regulate a technology it has tried not to regulate…
AI 点评 · 白宫转向开源模型监管，预示AI政策风向生变，值得关注。
可信度 74交叉信源 1
行业信号
8/12 17:51
At Ai4, three of the world's most respected AI experts — Geoffrey Hinton, Fei-Fei Li, and Andrew Ng — debated regulation, open source access, and how America can compete as China a…
AI 点评 · 三位AI泰斗罕见同台激辩开放与安全，为全球监管博弈提供关键风向标。
Skill / 资源
8/11 23:48
There are no lossless transformations of natural-language text Sophie Alpert shares her "internal policy on acceptable use of AI writing by engineers". It's a short read (supportin…
可信度 74交叉信源 1
论文 / 方法
8/11 20:00
Recent work on distillation transfers the capabilities of large models to smaller ones often by updating the latter's parameters, through teacher forcing, on-policy distillation, and related training-…
论文 / 方法
8/11 20:00
Large language models (LLMs) achieve remarkable performance across natural language tasks, yet they are trained on static corpora and their knowledge quickly becomes outdated in a fast-changing world.…
可信度 74交叉信源 1
论文 / 方法一手源
8/11 17:50
GUI Visual Grounding is a fundamental capability for GUI agents. Existing models typically freeze their parameters after deployment, limiting their ability to adapt to unseen interfaces. Although rece…
论文 / 方法一手源
8/11 16:18
When a tool-using agent is given the same task in a different language, does it still take the same steps? Multilingual evaluation rarely asks: it compares final answers and discards the actions. Yet…
论文 / 方法
8/10 20:00
We study reference-free post-training for multilingual machine translation with open large language models. Starting from the supervised-finetuned MiLMMT-46-v0.1 models, we apply Group Relative Policy…
论文 / 方法一手源
8/10 17:33
Financial prediction typically relies on task-specific regression, ranking, or policy heads, separating the language model from the numerical object ultimately evaluated. We investigate whether a caus…
行业信号
8/10 15:46
IT之家 8 月 10 日消息，据路透社报道，Meta CEO 马克 · 扎克伯格呼吁美国 减少开源 AI 面临的政策障碍 ，以便更好地与中国竞争对手抗衡。 今天早些时候，Meta 发布了新的开放权重模型 Muse Glimmer，更多模型很快也将推出。与竞争对手的顶尖 AI 模型相比，Muse Glimmer 规模更小，定位也有所不同。模型可以在 只有一张…
AI 点评 · 开源AI竞争升级，政策松绑成美国科技巨头核心诉求。
论文 / 方法一手源
8/10 15:40
On-policy self-distillation (OPSD) converts feedback into dense token-level supervision on trajectories generated by the policy to be optimized, providing a useful complement to reinforcement learning…
AI 点评 · 用自参考策略蒸馏把稀疏反馈变稠密监督，直击RL训练效率痛点，是提升模型稳定性的创新巧思。
行业信号
8/9 14:30
AI agents are escaping cybersecurity testing environments and reaching real-world systems, raising questions about whether safety infrastructure, industry standards, and regulation…
AI 点评 · 安全测试失控，AI代理逃逸暴露监管真空，安全防线反成风险源。
论文 / 方法
8/8 20:00
On-policy (Self-)Distillation (OPD / OPSD) has shown strong potential for post-training large language models (LLMs). However, existing methods still rely heavily on external supervision, including gr…
模型 / Agent
8/8 04:36
Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that frames content moderation as a single yes/no question instead of a fi…
论文 / 方法一手源
8/7 17:17
Wasserstein policy gradient (WPG) updates state-conditional action laws by transport in the action space. We study entropy-regularized discounted linear-quadratic (LQ) control. A Bellman verification…
论文 / 方法一手源
8/6 17:52
Multilingual reasoning transfer is crucial for extending reasoning capabilities of large language models (LLMs) beyond high-resource languages. On-policy self-distillation (OPSD) and its variants have…
行业信号
8/6 17:39
Suno announced plans to implement a new watermarking technology and download policy to limit the spread of spammy AI tracks and increase transparency. In a lengthy blog post, CEO a…
AI 点评 · AI音乐泛滥成灾，Suno用技术加水印治理，行业透明度有望提升。
Skill / 资源
8/6 16:43
Learn about new capabilities in Amazon Bedrock AgentCore: temporal policies powered by Dogwood, a new open source policy language for AI agents, and rate limiting on the gateway. T…
Skill / 资源
8/6 16:21
A regulated customer needed all Claude Code inference processed in a single AWS Region (London), not just in-geography. This post shows two ways to pin Claude Code on Amazon Bedroc…
Skill / 资源
8/6 16:12
Learn how to run the full Amazon Bedrock Automated Reasoning policy lifecycle from your coding agent. A suite of open source Agent Skills builds, reviews, tests, debugs, deploys, a…
行业信号
8/6 14:00
Today, I’m talking with Gaby Del Valle, a policy reporter here at The Verge, about the growing backlash against AI data centers. Gaby recently reported a fantastic piece about Hern…
论文 / 方法
8/5 20:00
On-Policy Distillation (OPD) is emerging as a promising alternative to reinforcement learning for LLM post-training, yet its effectiveness in multilingual settings remains underexplored. We study OPD…
论文 / 方法
8/5 20:00
Vision-Language-Action (VLA) models have become a powerful paradigm for robot manipulation, but training a single generalist policy for heterogeneous robot embodiments remains an open problem. Existin…
行业信号
8/5 11:43
Some creators fear the EU AI Act’s regulatory chaos will upend their lucrative businesses. Others are owning it by incorporating AI transparency into their creative process.
行业信号
8/5 05:52
论文 / 方法
8/4 20:00
Leading open text-to-image models often carry complementary strengths: one may lead on preference-aligned aesthetics while another follows compositional instructions more faithfully. However, differen…
论文 / 方法
8/4 20:00
On-Policy Self-Distillation (OPSD) has become a standard post-training approach for improving visual reasoning in multimodal large language models (MLLMs). Existing methods draw privileged information…
论文 / 方法
8/4 20:00
Once visual content enters an AI pipeline, its owner often retains little technical control over how it is used. Legal and regulatory remedies can address misuse, but many technical interventions must…
论文 / 方法
8/4 20:00
Privileged on-policy distillation provides dense supervision for multi-turn agents by allowing a synchronized teacher to re-score the student's response at every turn with access to training-only refe…
论文 / 方法
8/4 20:00
On-policy distillation (OPD) provides dense teacher supervision on student-generated trajectories, but standard reverse-KL training can assign insufficient probability to other plausible continuations…
论文 / 方法一手源
8/4 17:34
Generative Flow Networks (GFlowNets) have emerged as a flexible framework for amortised inference over discrete and mixed discrete-continuous objects, requiring only an unnormalised target density spe…
行业信号
8/4 15:42
Tech companies and developers have been scouring the U.S. for places to build data centers, and they’ve been drawn to Texas’ loose regulations and seemingly abundant power supply.…
行业信号
8/3 23:08
欧盟《人工智能法案》的核心透明度条款正式生效，标志着全球首部全面监管人工智能的法律进入实质执行阶段。根据新规，聊天机器人等交互式系统必须明确告知用户其正在与人工智能而非人类互动；利用AI生成或修改的图像、视频、音频及文本内容，须进行清晰标识并附带机器可读的数字标记。（新浪财经）
AI 点评 · 全球首部全面AI法案落地，透明度规则将重塑行业合规标杆。
论文 / 方法
8/3 20:00
On-policy training has emerged as a powerful post-training paradigm for improving the reasoning capabilities of large language models, and is often enhanced by golden trajectories from stronger expert…
论文 / 方法
8/3 20:00
On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, a…
论文 / 方法
8/3 20:00
On-Policy distillation (OPD) transfers teacher capabilities by supervising student-sampled trajectories with dense token-level teacher signals. Recent selective OPD methods improve this process by pri…
论文 / 方法一手源
8/3 17:51
World Action Models (WAMs) augment robot policies with action-conditioned predicted futures, but a plausible future alone does not justify changing the action that a bimanual policy would execute. We…
论文 / 方法一手源
8/3 17:37
Cognitive AI seeks to move beyond language generation and autonomous task execution toward systems capable of sustained reasoning, adaptive behavior, persistent memory, and self-regulation. While gene…
Skill / 资源
8/3 16:30
Amazon Bedrock now supports automatic Automated Reasoning policy refinement. The refinement engine diagnoses failing tests and proposes formal-logic fixes for rule issues and langu…
AI 点评 · 自动修复策略缺陷，大幅降低AI规则调试门槛。
论文 / 方法
8/2 20:00
On-policy (self) distillation (OPSD) is increasingly adopted for language-model post-training. It strengthens the teacher with privileged information but can induce a privilege illusion: the student l…
论文 / 方法
8/1 20:00
World Action Models (WAMs) couple action generation with prediction of future states. Their effectiveness depends on whether future dynamics are modeled in a space that is both aligned with action gen…
论文 / 方法
7/31 20:00
Reinforcement learning with verifiable rewards (RLVR) has become a standard paradigm for post-training large language models (LLMs). While Group Relative Policy Optimization (GRPO) is widely adopted,…
论文 / 方法一手源
7/31 16:52
Imitation learning (IL)---training an agent to replicate expert behavior from demonstrations---underpins applications from robotics to language model training. Standard approaches such as Behavior Clo…
论文 / 方法一手源
7/31 16:38
A fundamental challenge of vector search is achieving consistently high recall while minimizing computational costs. Fixed search parameters cause significant performance variance across queries, and…
论文 / 方法一手源
7/31 16:33
Parameter-Efficient Fine-Tuning (PEFT) commonly adapts large language models using a single shared Low-Rank Adapter (LoRA). This shared optimization space often suffers from interference when adapting…
论文 / 方法一手源
7/31 16:19
This paper studies the policy gradient update for a multi-arm bandit problem in diffusion environment that is described by a stochastic differential equation (SDE) under the continuous-time reinforcem…
行业信号
7/31 12:59
36氪获悉，7月31日，国务院国资委党委召开扩大会议。会议强调，要指导中央企业主动对接和带头落实好国家政策，抓住重点、主动作为，充分发挥引领带动作用，更好助推经济持续向新向优向好发展。要锚定全年目标不动摇，切实加强现场督导调研，把中央企业生产经营情况摸清摸透，找准痛点、对症下药，夯实企业高质量发展基础。要着力推进原始创新和关键核心技术攻关，以更大力度加强基础…
行业信号
7/30 23:11
Meta表示，已通过长期和短期协议，承诺了近7000亿美元的未来支出，涉及人工智能数据中心、云计算等领域。Meta在周四的一份监管申报文件中表示，公司拥有3493亿美元的不可撤销合同承诺，主要涉及第三方云服务协议、服务器和网络基础设施。该公司还有3470亿美元尚未开始的租赁承诺，尚未反映在资产负债表中。其中仅7月份就新增了680亿美元，付款将从2027年和2…
AI 点评 · 巨额投资揭示科技巨头对AI基础设施的豪赌，预示行业竞争将更激烈。
行业信号一手源
7/30 20:35
MIT students and postdocs discussed science funding and research with policymakers in Washington during the MIT Science Policy Initiative’s annual Congressional Visit Days.
AI 点评 · 学生与政策制定者直接对话，推动科研经费议题，展现青年影响决策的实践路径。
行业信号一手源NEW
7/30 20:35
MIT students and postdocs discussed science funding and research with policymakers in Washington during the MIT Science Policy Initiative’s annual Congressional Visit Days.
可信度 88交叉信源 1
论文 / 方法
7/30 20:00
Reinforcement learning with verifiable rewards (RLVR) broadcasts a single response-level reward to every token, while on-policy distillation (OPD) scores each token against a stronger teacher for a de…
论文 / 方法一手源
7/30 17:59
We present PAC-MAN, a perception-aware CBF-RL framework that couples control-barrier safety with deployment-realistic onboard sensing for whole-body humanoid dodgeball. The deployed policy sees the ba…
Skill / 资源
7/30 17:51
This week on Uncanny Valley, we discuss the open- vs. closed-source debate in AI, key players in White House AI policy, and how to stop your chatbot logs from showing up in search-…
AI 点评 · 英伟达开源联盟缺失OpenAI和Anthropic，揭示AI巨头间开源与闭源路线的深层博弈。
论文 / 方法一手源
7/30 17:43
Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-…
论文 / 方法一手源
7/30 17:21
Predicting the 3D structures of atomic systems is fundamental to advancing material science and drug discovery. While flow-matching models (, FlowDPO) have recently shown promise in this domain, their…
行业信号
7/30 14:53
Plus, a new policy for the AI protocol ensures features aren't removed suddenly.
行业信号
7/30 13:00
The Series A was led by Khosla Ventures, with participation from Allianz, Rebel Fund, Brick and Mortar Ventures’ Darren Bechtel, and Y Combinator’s Garry Tan.
行业信号
7/30 12:36
行业信号
7/30 11:45
行业信号
7/29 23:14
36氪获悉，近日，上海市市场监督管理局制定发布《上海市户外广告发布内容合规指引》。针对“大字吸睛、小字免责”等问题，《合规指引》要求提示性用语真实、清晰、显著，重要交易条件和限制性信息应当完整准确提示；规范使用“第一”“首创”“最佳”“领先”等用语及广告引证内容，不得实施“萝卜坑式引证”；使用人工智能生成合成内容开展广告宣传的，应当依法履行标识义务。
AI 点评 · AI广告须标明身份，上海新规填补监管空白，推动技术透明与消费者知情权。
论文 / 方法
7/29 20:00
On-policy self-distillation (OPSD) is a promising approach to improve reasoning language models, but it remains brittle in practice: making it work reliably often requires substantial engineering effo…
论文 / 方法
7/29 20:00
Large language model training in open-ended domains lacks verifiable rewards, making task preferences difficult to formalize as effective supervision. Contexts can convey such preferences, yet provide…
论文 / 方法
7/29 20:00
Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-…
行业信号
7/29 14:22
IT之家 7 月 29 日消息，近日，上海市市场监督管理局制定发布《上海市户外广告发布内容合规指引》（以下简称《合规指引》），系统梳理户外广告发布中的重点风险和合规要求，为广告活动主体加强内容审核、防范违法违规风险提供更加清晰、具体、可操作的指引。 针对“大字吸睛、小字免责”等问题，《合规指引》要求提示性用语真实、清晰、显著，重要交易条件和限制性信息应当完整…
行业信号
7/29 12:40
7月29日从业内获悉，上海证券交易所理事会投资者教育和保护专门委员会今年确定三项重点研究课题——非法证券活动的防控机制建设、人工智能在证券服务中的应用边界与合规治理、证券公司投诉典型情形识别与纠纷化解机制。目前，对应三份专项调研问卷已陆续发至各会员单位，一场覆盖证券行业的投保体系“深度体检”正在展开。券商人士表示，从问卷设计来看，监管思路正从“有没有”转向“…
论文 / 方法一手源
7/28 17:59
Generalist manipulation policies increasingly take the form of action-chunking flow policies built on large pretrained backbones. Such chunks run open-loop, so the policy cannot react to sensory input…
行业信号
7/28 14:58
IT之家 7 月 28 日消息，据《政治报》昨天报道，OpenAI 首席执行官萨姆 · 奥尔特曼（Sam Altman）预计将于周三前往华盛顿，与白宫官员和国会两党议员举行会谈，讨论美国的人工智能政策。 据消息人士透露，奥尔特曼此行预计将与美国财政部长斯科特 · 贝森特（Scott Bessent）、美国商务部长霍华德 · 卢特尼克（Howard Lutni…
论文 / 方法
7/27 20:00
On-policy distillation (OPD) grounds token-level supervision in the student's own trajectory, yet suffers from prefix failure: once the student commits to a wrong reasoning direction, all subsequent g…
论文 / 方法
7/27 20:00
We introduce Shieldstral, a 3B-parameter policy-adaptive multimodal safety classifier that matches or outperforms models nearly 7times its size on text safety benchmarks and sets a new state of the ar…
论文 / 方法
7/27 20:00
Generalist manipulation policies increasingly take the form of action-chunking flow policies built on large pretrained backbones. Such chunks run open-loop, so the policy cannot react to sensory input…
论文 / 方法
7/27 20:00
Rubric-based reinforcement learning enriches language model training by evaluating model outputs against explicit criteria. Yet in GRPO-style pipelines, these structured judgments are reduced to a sca…
论文 / 方法
7/27 20:00
On-policy distillation (OPD), which aligns a student with the teacher's token-level distribution on the student's own rollouts, is an effective paradigm for transferring capabilities across LLMs. Prev…
论文 / 方法一手源
7/27 17:55
Multi-turn long-horizon planning is critical for foundation model agents, yet how to fundamentally improve it remains unclear. Existing models are trained on uncontrollable and opaque Internet data, m…
AI 点评 · 探索从预训练到后训练提升大模型长程规划能力，为自主智能体发展提供关键路径。
论文 / 方法一手源
7/27 17:14
In safety-critical sectors such as robotics and automotive engineering, the deployment of Deep Reinforcement Learning (DRL) is often hindered by the black-box nature of deep neural networks. This lack…
AI 点评 · 用物理知识蒸馏强化学习策略，兼顾性能与可解释性，推动安全关键领域应用。
行业信号
7/27 10:00
“It’s not an argument with two sides, it’s an argument with 10 sides,” one senior administration official tells WIRED about how US AI policy is being shaped.
论文 / 方法
7/26 20:00
On-policy distillation (OPD) adapts diffusion models by querying a teacher along trajectories generated by the current student, but how it should behave under classifier-free guidance (CFG), a default…
论文 / 方法
7/26 20:00
Multi-turn long-horizon planning is critical for foundation model agents, yet how to fundamentally improve it remains unclear. Existing models are trained on uncontrollable and opaque Internet data, m…
论文 / 方法
7/26 20:00
Biomedical image analysis spans diverse modalities and tasks, yet real-world deployment is hindered by severe distribution shifts across scanners, protocols, and patient populations. High-performing m…
行业信号
7/24 15:51
AI companies, including Nvidia and Mistral, urge policymakers to avoid broad restrictions on open-weight AI models as Washington debates responses to Chinese AI and alleged model d…
行业信号
7/24 15:15
IT之家 7 月 24 日消息，谷歌宣布签署欧盟《人工智能法案》框架下的《AI 生成内容透明度行为准则》 ，以支持在欧洲市场负责任地使用人工智能。 据悉，该准则核心要求主要集中在以下四个方面： 机器可读的标识与水印：确保第三方工具或平台能够通过技术手段自动检测并识别该内容是由 AI 生成或修改的； 人机交互的明确告知：当 AI 系统直接与人类互动时，必须在设…
行业信号
7/24 13:32
Letter: https://images.nvidia.com/pdf/Open-Weights-and-American-AI-L... [pdf] https://x.com/JensenHuang/status/2080643682408321103 , https://xcancel.com/JensenHuang/status/20806436…
设计 / 产品
7/24 11:00
大公司： 字节跳动在中卫成立朔方星河科技公司，注册资本24亿 36氪获悉，天眼查App显示，近日，中卫朔方星河科技有限公司成立，法定代表人为田凯，注册资本24亿人民币，经营范围包括信息技术咨询服务、软件开发、信息系统集成服务、网络设备销售等。股东信息显示，该公司由北京字跳网络技术有限公司全资持股。 周生生、老凤祥等品牌金饰价格集体下调 当地时间7月23日，黄…
论文 / 方法
7/23 17:40
Building socially calibrated large language models, which can learn from others without simply yielding to them, requires more than reducing sycophancy as a one-dimensional failure mode. Models must d…
AI 点评 · 突破单维谄媚缺陷，揭示LLM道德推理中结构化抵抗与服从的平衡机制。
论文 / 方法
7/23 17:37
On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to en…
AI 点评 · 自蒸馏无需外部教师，简化流程并提升效率，是自监督学习的实用突破。
论文 / 方法
7/23 17:35
While large audio-language models have achieved remarkable progress in auditory perception, they still lag behind text-based large language models in deep logical reasoning, primarily due to the scarc…
AI 点评 · 用强化学习对齐策略，让音频大模型学会逻辑推理，突破听觉感知瓶颈。
行业信号
7/23 11:08
AI 点评 · 聚焦智能体政策动向，解读AI治理新趋势，关乎行业合规发展。
论文 / 方法
7/22 20:00
On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to en…
AI 点评 · 自监督蒸馏新范式，无需外部教师，突破信息不对称瓶颈。
论文 / 方法
7/21 20:00
Reinforcement learning with verifiable rewards has become the predominant recipe for eliciting test-time scaling in explicit Chain-of-Thought reasoners. Yet this scaling path remains computationally c…
AI 点评 · 提出替代策略提升潜在推理效率，为降低AI推理成本提供新思路。
论文 / 方法
7/21 20:00
In RLHF pipelines, reward scoring blocks policy updates. Slow scoring bottlenecks the entire loop, since no update runs until every rollout gets a score. And yet most setups just default to PyTorch ea…
论文 / 方法
7/20 20:00
Modern ASR models trained on heterogeneously annotated data treat transcription style (verbatim vs. intended) as an uncontrolled latent variable, causing measurable decoding instability, evaluation co…
AI 点评 · 用词级时间戳激活可控逐字识别，解决ASR转录风格不一致导致的解码不稳定问题。
论文 / 方法
7/20 20:00
Asynchronous reinforcement learning improves throughput by decoupling rollout generation from optimization, but staleness is an inevitable byproduct compounded by policy lag, engine delays, and mixtur…
论文 / 方法
7/20 17:59
Pretrained dense visual features from Vision Transformers (ViTs) are powerful yet have been underutilized in robot learning. Modern robot policies either compress each observation into a single global…
论文 / 方法
7/20 17:57
We propose one-step and two-step methods for policy learning with retrieval-augmented generation (RAG). We formulate RAG-based action selection under the potential outcome framework. In the two-step m…
论文 / 方法
7/20 17:07
PPO and the GRPO baseline studied here use clipped surrogate objectives whose favorable-direction saturation introduces an abrupt change in the scalar objective's derivative. We ask whether Output Res…
论文 / 方法
7/20 12:31
Welcome to Import AI, a newsletter about AI research. Import AI runs on arXiv, cappuccinos, and feedback from readers. If you’d like to support this, please subscribe. Subscribe no…
可信度 74交叉信源 1
论文 / 方法
7/19 20:00
We propose Token-Level Off-Policy Labeling (TOPL), an off-policy training paradigm that reframes post-training as a token-level correctness prediction task. Our key intuition is that by training the m…
论文 / 方法
7/18 20:00
Large language model (LLM) post-training is essential for improving reasoning, adaptation, and alignment. Existing methods mainly follow two paradigms: reinforcement learning (RL) and on-policy distil…
行业信号
7/18 14:30
IT之家 7 月 18 日消息，据路透社报道，反对 AI 数据中心快速扩张的民众计划于周六（7 月 18 日）在全美至少 125 个地点举行抗议活动。 此次抗议活动由一个名为“人类优先 (HumansFirst)”的民间组织牵头。抗议者将举行集会，反对“不受控的”数据中心扩张行为，以及“对我们自由不可容忍的侵犯”。 尽管遭到居民的强烈反对或缺乏监管审查，报道…
AI 点评 · AI扩张引发民间反弹，折射出科技发展与社区权益的深层矛盾。
论文 / 方法
7/17 20:00
Entropy control has become an effective tool in reinforcement learning (RL) of large language models (LLMs), helping balance exploration-exploitation trade-off during alignment process. Such RL paradi…
论文 / 方法
7/17 20:00
On-policy knowledge distillation transfers reasoning from large teachers to compact students, but existing approaches suffer three compounding failure modes: (i) cold-start collapse, where a fresh stu…
论文 / 方法
7/17 16:20
Transferring policies across domains poses a vital challenge in reinforcement learning, due to the dynamics mismatch between the source and target domains. In this paper, we consider the setting of on…
行业信号
7/16 18:35
The company endorsed landmark AI transparency laws in California and New York last year, but its head of US state and local policy says they may already be outdated.
AI 点评 · Anthropic推动加速AI监管，显示行业巨头对现有法规可能滞后于技术发展的紧迫感。
论文 / 方法
7/16 17:45
Evidence synthesis is crucial for turning primary research into reliable knowledge for science, medicine, education, and policy. Yet, quantitative evidence synthesis remains largely manual and difficu…
AI 点评 · AI自动化元分析系统，极大提升科研效率，降低人工成本，推动循证决策发展。
论文 / 方法
7/16 17:31
We develop data-driven algorithms for maintaining $N$ independent identical machines under a \textit{block replacement policy}, in which each machine is replaced upon failure and all machines are join…
AI 点评 · 用数据驱动替代固定周期换件，显著降低维护成本，革新传统设备管理策略。
论文 / 方法
7/15 20:00
Large language models are increasingly trained as interactive agents for long-horizon tasks involving multi-turn interaction, tool use, and environment feedback. Outcome-based reinforcement learning (…
论文 / 方法
7/15 20:00
Reinforcement learning with verifiable rewards (RLVR) commonly uses entropy for advantage shaping. However, entropy cannot distinguish useful uncertainty from detrimental confusion, limiting its effec…
论文 / 方法
7/15 20:00
On-policy distillation is an alternative post-training method in reinforcement learning that alleviates the constraints imposed by reward models by providing token-level supervision from a teacher mod…
论文 / 方法
7/15 20:00
We study on-policy distillation (OPD) for agentic tasks, where an LLM agent interacts with an environment over multiple turns and a student imitates a teacher over these multi-turn interaction histori…
论文 / 方法
7/14 20:00
World Action Models (WAMs) improve robot policy learning by jointly modeling actions and future visual observations, using future scene evolution as dense supervision for physically grounded action ge…
论文 / 方法
7/14 20:00
On-policy distillation (OPD) has become a key paradigm in LLM post-training, yet its training dynamics remain poorly understood. We present a systematic study examining the role, pathologies, and regu…
论文 / 方法
7/14 20:00
Validating autonomous driving systems requires diverse, regulation-compliant test scenarios. In simulation-based testing, scenarios are defined as executable scripts. Yet automatically generating such…
论文 / 方法
7/14 20:00
Agentic language models must learn when to call tools, when to consume tool responses, and when to answer directly. This makes multi-teacher on-policy distillation a natural training strategy: one tea…
行业信号
7/14 17:45
DeepMind CEO Demis Hassabis is proposing an AI "standards body" modeled after FINRA, to test frontier models and develop best practices for their release.
AI 点评 · 呼吁建立独立标准机构监管前沿AI，体现行业领袖对安全治理的紧迫感与务实路径。
行业信号
7/14 15:17
IT之家 7 月 14 日消息，据科技媒体 The Verge 今天报道，谷歌 DeepMind 创始人兼 CEO 德米斯 · 哈萨比斯（Demis Hassabis）认为，美国政府应成立全球 AI 监督机构， 让人们能够在前沿 AI 模型发生险情时踩下“刹车” 。 哈萨比斯认为，美国应该主导这一倡议并制定全球标准。他认为这个设想中的机构可以参考美国金融业监…
行业信号
7/14 10:48
凌晨三点，一家刚成立不久的AI创业公司，可能已经在同时服务旧金山的客户、采购首尔的技术服务，并与拉各斯的合作伙伴签下合同。这家公司甚至还没有招到第一名全职财务人员，业务却已经跨越多个市场、币种和监管辖区。 AI正在让这样的创业路径成为可能。过去需要市场、运营、客服等一整套全球化团队才能完成的工作，现在借助智能体就能承担相当一部分。新一代初创企业不必再按照“先…
行业信号
7/14 09:00
In response to a public records request, HUD has withheld documents about DOGE’s use of AI—in part by citing a privilege that doesn’t exist.
行业信号
7/14 00:04
今日热点导览 韩国法院禁止两名前三星员工就职于SK海力士 字节探索自动驾驶，Seed世界模型团队负责 巨力索具深夜澄清商业航天业务不实传闻，提醒投资风险 科技记者古尔曼：苹果将跳过M6 Pro/Max芯片全力押注AI导向的M7系列 美众议院本周将表决永久实行夏令时法案 TOP 3大新闻 海南将成中国首个禁售燃油车省份，届时车桩比保持2.5:1以下 近日，海南…
论文 / 方法
7/13 20:00
Structured pruning is a hardware-friendly way to compress LLMs, but it is mostly validated on multiple-choice recognition tasks, while the same compressed checkpoints can collapse on the free-form gen…
论文 / 方法
7/13 16:28
Explainability has emerged as a critical requirement for AI-based systems, particularly in safety-critical and regulated domains. Although prior research has proposed frameworks, patterns, and user-ce…
论文 / 方法
7/13 08:56
Post-training is essential for refining the domain-specific capabilities of large language models (LLMs), yet existing reward optimization and distribution matching methods tightly couple policy explo…
论文 / 方法
7/11 20:00
Reinforcement learning for large language models (LLMs) typically relies on trust-region masks to stabilize off-policy updates. The dominant PPO-style approach uses the sampled-token importance ratio…
行业信号
7/11 14:58
IT之家 7 月 11 日消息，近期有消息称，腾讯正考虑入股 AI 初创公司 Manus，计划成为该公司的最大股东。腾讯正与真格基金、红杉资本和 Manus 管理层磋商，计划按 20 亿美元收购这家 AI 公司。 据财闻今晚报道，接近交易的知情人士表示， 此次 股权变动 是在 监管主导之下 ， 社会资本共同补齐此前 Meta 的投资，以便让 Meta 退出…
AI 点评 · 腾讯入股Manus或为监管主导下的资本重组，Meta退出值得关注。
论文 / 方法
7/10 20:00
Agentic retrieval-augmented generation (RAG) extends static RAG by allowing language models to iteratively reason, generate search queries, retrieve evidence, and predict answers. However, it remains…
论文 / 方法
7/10 20:00
Reinforcement learning (RL) has become a dominant paradigm for enhancing LLMs' reasoning capabilities. However, RL algorithms with PPO-Clip are inherently limited by exploration collapse. Subsequent w…
论文 / 方法
7/10 16:27
Pre-demolition assessment, the regulated audit process at the heart of urban mining, is an information process in which AI support must serve qualified auditors who remain accountable for the decision…
AI 点评 · 知识图谱与可解释AI结合，让城市矿产审计更透明可信。
模型 / Agent
7/9 17:00
About two weeks after OpenAI's GPT-5.6 was caught up in regulatory drama - rolled out only to government-approved organizations during a "limited preview" period - the company has…
论文 / 方法
7/8 20:00
We propose OPSD-V, an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models. Existing few-step AR video generators can produce long videos with low…
模型 / Agent
7/8 19:49
Today, we're announcing the Claude apps gateway for AWS, a self-hosted control plane that gives organizations a single point of control over access, cost, and policy for Claude Cod…
AI 点评 · 企业级AI部署新方案，统一管理Claude访问权限与成本，解决安全合规痛点。
论文 / 方法
7/8 17:32
Group Relative Policy Optimization (GRPO) stalls on a model's hardest problems: when no rollout in a group succeeds, the group-relative advantages vanish and the problem contributes no gradient, wasti…
模型 / Agent一手源
7/8 07:00
Anthropic’s Responsible Scaling Policy Anthropic
论文 / 方法
7/7 20:00
Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must…
模型 / Agent
7/6 23:27
鉴于ChatGPT、Claude和Gemini等大型语言模型对消费者财务决策的影响日益增加，英国金融监管机构被敦促考虑对这些模型进行监管。在英国金融市场行为监管局（FCA）委托进行、于周一发布的一份评估报告中，FCA执行董事米尔斯（Sheldon Mills）也强调，企业对少数几家技术供应商的依赖可能带来全系统范围的风险。（新浪财经）
AI 点评 · 监管关注金融领域AI依赖少数巨头，防范系统性风险与消费者保护成关键看点。
论文 / 方法
7/6 20:00
On-policy distillation (OPD) trains a student policy by matching a stronger teacher on the student's own trajectories, offering a promising framework for language agent training. However, its applicat…
论文 / 方法
7/6 20:00
Image guardrails are typically trained and evaluated under a fixed safety policy, implicitly treating safety as an intrinsic property of an image. Real deployments are different: the same image may be…
论文 / 方法
7/6 17:59
Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must…
论文 / 方法
7/6 17:53
Occupancy ratios correct distribution shift in offline reinforcement learning and are central to off-policy evaluation. Existing primal-dual and minimax methods typically estimate these ratios by enfo…
论文 / 方法
7/6 17:21
Group Relative Policy Optimization (GRPO) is effective when the current policy already samples useful reasoning trajectories, but it stalls on hard prompts whose correct solution modes lie outside the…
行业信号
7/6 14:17
FCA official makes case for greater powers for watchdog as millions use technology for personal finance decisions.
论文 / 方法
7/5 20:00
For robots to work reliably in commercial and industrial applications, can recent advances in agentic coding systems combine interpretable robot programming with the open-world adaptability of model-f…
论文 / 方法
7/5 20:00
Unified models for robot manipulation aim to equip one policy with both the semantic priors of pretrained VLMs and the physical dynamics learned through future prediction. In practice, existing design…
论文 / 方法
7/5 20:00
Group Relative Policy Optimization (GRPO) is effective when the current policy already samples useful reasoning trajectories, but it stalls on hard prompts whose correct solution modes lie outside the…
论文 / 方法
7/5 20:00
Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust Region Policy Distillation (TOP-D), which transforms the notoriously unstable, high-variance On-Pol…
论文 / 方法
7/4 20:00
Diffusion large language models (dLLMs) generate text by iteratively denoising a masked sequence, offering a parallel alternative to autoregressive models, but eliciting strong reasoning through post-…
论文 / 方法
7/4 20:00
Recent advances in multimodal foundation models and agent systems have driven GUI agents from single-platform task execution toward cross-platform interaction. However, building multi-platform GUI age…
论文 / 方法
7/4 20:00
Reinforcement learning (RL) for non-verifiable instruction following increasingly relies on LLM judges with prompt-specific rubrics as reward signals. While recent methods adapt these rubrics to the e…
行业信号
7/4 01:03
今日热点导览 三部门：调整节能汽车、新能源汽车车船税优惠政策 三星传获Meta超10万亿韩元AI芯片代工订单 茉莉奶白小程序更换彩色Logo Meta打算出售富余算力引发科技股回落 英伟达前光互连技术高管Ashkan Seyedi加入艾迈斯欧司朗 TOP3大新闻 因存在植入后门风险，阿里内部全面禁用Claude Code 36氪从阿里内部人士处获悉，因近期C…
AI 点评 · 阿里禁用Claude Code凸显企业数据安全红线，AI工具合规性成焦点。
论文 / 方法
7/3 20:00
Visual policies learned from human videos, teleoperation, and robot demonstrations offer scalable motion priors, but often fail in contact-rich manipulation, where success significantly depends on loc…
设计 / 产品
7/3 14:16
IT之家 7 月 3 日消息，豆包今晚发布《豆包智能体功能下线通知》，称由于产品功能调整， 智能体功能将于 2026 年 7 月 15 日下线 。 《通知》显示，该功能下线后，用户仍可在一段时间内查看并自行保存智能体信息及历史对话数据。2026 年 10 月 15 日后，豆包将根据《隐私政策》对智能体相关数据进行处理， 后续将无法在豆包内查看或恢复 。如有重…
AI 点评 · 产品功能调整背后，需关注用户数据迁移与隐私政策变化对AI服务稳定性的影响。
论文 / 方法
7/2 17:58
On-policy self-distillation (OPSD) has emerged as a practical method for training large language models (LLMs) to reason, where a single model acts as both the teacher and the student with different l…
论文 / 方法
7/1 20:00
Autonomous agents are increasingly expected to improve executable policies through feedback, yet existing evaluations often collapse this process into a final score or confound it with open-ended soft…
论文 / 方法
7/1 20:00
Continual post-training enables foundation models to acquire new knowledge while preserving existing capabilities. Recent work suggests that on-policy learning can mitigate forgetting, with on-policy…
论文 / 方法
7/1 20:00
Vision-Language-Action (VLA) foundation models have recently achieved strong progress in embodied intelligence. To reduce policy-call frequency while preserving temporal coherence, most generative pol…
论文 / 方法
7/1 20:00
We present EVA-Client, an open-source framework for deployment, data collection, and evaluation of trained manipulation policies on real robots. Sitting between a policy server and the physical hardwa…
论文 / 方法
7/1 20:00
Evaluating embodied robot foundation models remains a critical bottleneck; unlike large language models efficiently assessed via digital benchmarks, robotic policies require slow, costly real-world ro…
行业信号
7/1 17:48
Cloudflare is giving AI companies until September 15 to separate web crawlers used for search from those used for AI training and agents, or risk being blocked by default on many p…
AI 点评 · 云服务商首次明确要求AI训练爬虫付费，或重塑数据获取规则。
行业信号
7/1 02:16
The Trump administration's erratic approach to AI policymaking has left companies across the industry with little clarity about what will govern future model releases.
论文 / 方法
6/30 20:00
Touch supplies the physical grounding needed to perceive intrinsic material properties, such as friction and compliance, that vision alone often cannot resolve. Recent efforts for equipping multimodal…
论文 / 方法
6/30 17:54
Reward design remains a central bottleneck for autonomous robot policy improvement, especially in long-horizon manipulation tasks where sparse success labels provide too little signal and binary prefe…
论文 / 方法
6/30 17:37
Policy-grounded document review requires determining whether a target document complies with organization-specific policies, guidelines, or playbooks. While large language models can assist with polic…
论文 / 方法
6/29 20:00
Metacognition is a critical component of intelligence that describes the ability to monitor and regulate one's own cognitive processes. Yet LLMs exhibit systemic deficiencies in key metacognitive facu…
论文 / 方法
6/29 20:00
Speculative decoding accelerates inference by using a lightweight draft model to generate candidate tokens in parallel, and are then verified by the target model, enabling lossless acceleration. Recen…
论文 / 方法
6/29 17:56
Conservative offline training is widely advocated as a safe foundation for subsequent online adaptation: if a policy stays close to well-supported behaviour, the argument goes, it is less likely to ex…
论文 / 方法
6/29 17:55
On-policy distillation (OPD) offers superior capacity transfer by supervising student-sampled trajectories with dense token-level signals. To furnish high-quality supervision sources and thereby eleva…
设计 / 产品
6/29 14:05
Open-source local policy, recovery, and audit layer for explicit Codex execution.
行业信号
6/29 13:09
论文 / 方法
6/28 20:00
In real-world applications, guardrails are often expected to identify unsafe user-model interactions according to application-specific safety policies, rather than relying on predefined risk taxonomie…
论文 / 方法
6/28 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 metho…
论文 / 方法
6/28 20:00
On-policy distillation (OPD) offers superior capacity transfer by supervising student-sampled trajectories with dense token-level signals. To furnish high-quality supervision sources and thereby eleva…
论文 / 方法
6/27 20:00
LLM agents handle user requests on behalf of organizations through tool calls and must follow the company policies stated in their system prompts. Prior work approaches this as a safeguarding problem…
设计 / 产品
6/27 09:21
6月26日，北京人形机器人创新中心慧思开物平台的双大脑模型天鹕（Pelican-VL）和我悟（WoW）同步完成北京市网信办最新一批生成式人工智能服务备案。北京人形将正式启动慧思开物全系列模型Token服务，计划分阶段面向产业客户、科研机构、开发者全面开放API调用能力。（界面新闻）
AI 点评 · 具身智能迈入备案监管时代，标志人形机器人商业化落地加速。
设计 / 产品
6/26 23:10
IT之家 6 月 27 日消息，据央视新闻 6 月 25 日报道，市场监管总局正会同相关部门，加快智能体等前沿技术领域标准制定速度，动态完善适配产业发展的人工智能国家标准矩阵。 报道称，目前正在抓紧制定的国家标准，除智能体外，还有 具身智能、世界模型、本体模型 等前沿技术标准，算力基础设施、高质量数据集、仿真测试平台、深度学习编译器、开源模型平台等底座类标准…
AI 点评 · 政策加速标准制定，将推动智能体与具身智能产业规范化发展，抢占技术制高点。
模型 / Agent
6/26 17:00
Less than 24 hours after news broke that OpenAI would stagger its next model release at the request of the Trump administration, that model, GPT-5.6, is here. On Friday, the compan…
AI 点评 · 政商博弈下加速发布，凸显AI巨头在监管压力中的技术突围与市场策略。
Skill / 资源
6/26 14:38
In this post, you learn how Stripe built a production-grade AI agent system for financial compliance. We cover the technical architecture of Stripe’s ReAct agent framework and the…
行业信号
6/25 23:29
IT之家 6 月 26 日消息，据《纽约时报》昨日援引三位内部人士消息，OpenAI 正考虑将 IPO（首次公开募股）推迟至明年进行。 此前曾有传闻称，OpenAI 已向美国监管机构秘密递交 IPO 申请，目标估值最高可达 1 万亿美元（IT之家注：现汇率约合 6.82 万亿元人民币）。但根据内部知情人士所述，目前 OpenAI 向顾问公司提出了两种方案：…
AI 点评 · 推迟IPO或为稳固万亿美元估值，反映其战略定力与市场博弈，值得关注。
论文 / 方法
6/25 20:00
Training and evaluating robot policies in the real world is costly and difficult to scale. We introduce SimFoundry, a modular and automated system for zero-shot real-to-sim scene construction from a v…
论文 / 方法
6/24 20:00
Outcome-based reinforcement learning provides a stable optimization backbone for language agents, but its sparse trajectory-level rewards provide little guidance on which intermediate decisions should…
论文 / 方法
6/24 20:00
Modern image generation demands a single model that unifies diverse capabilities, including text-to-image (T2I), local editing, and global editing. However, these capabilities are rarely naturally ali…
论文 / 方法
6/24 20:00
We present Qwen-Image-2.0-RL, a post-training pipeline that applies reinforcement learning from human feedback (RLHF) and on-policy distillation (OPD) to improve both the visual quality and instructio…
论文 / 方法
6/24 17:59
Most Vision-Language-Action (VLA) models build on a Vision-Language Model (VLM) backbone by attaching an action module and optimizing the full policy jointly. This design inherits strong visual and li…
论文 / 方法
6/24 17:59
On-policy self-distillation achieves strong pass@1 accuracy by using a single model as both teacher and student, with the teacher conditioned on a correct demonstration to provide dense token-level fe…
行业信号
6/23 22:56
IT之家 6 月 24 日消息，据纽约时报消息，知情人士透露，美国政府正向 Meta 施压， 要求其主动提交人工智能模型以供审查 ，以便政府能够评估这些模型的功能与潜在漏洞。 IT之家从报道获悉，在目前美国主要的 AI 技术开发商中， Meta 是唯一一家尚未与政府达成自愿共享模型以供审查协议的公司 。目前 OpenAI、Anthropic、谷歌、xAI 和…
AI 点评 · 监管力破AI巨头技术壁垒，Meta若松口将成行业风向标，安全审查或成全球治理新标杆。
论文 / 方法
6/23 20:00
Fine-grained visual reasoning requires multimodal large language models (MLLMs) to identify task-relevant visual evidence and ground their reasoning in local image regions. Existing agentic methods ty…
行业信号
6/23 17:00
Hello and welcome to Regulator, the newsletter for Verge subscribers chronicling the misadventures of their favorite tech overlords and Washington swamp creatures. ("Favorite" is,…
AI 点评 · 企业AI超级政治行动委员会重金投入地方选举，揭示科技巨头正深度介入政治权力布局。
行业信号
6/23 08:47
大公司： 甲骨文上财年裁员约2.1万 甲骨文上财年裁员约2.1万，减员规模大于此前已知水平，其中包括因使用人工智能（AI）而被取消的岗位。甲骨文周一在年度财务监管申报文件中称：“我们在各项业务中采用和部署AI技术，已经导致并且未来可能继续导致员工人数减少。” 英伟达宣布Vera Rubin NVL4系统Q4起供货 6月22日，英伟达宣布推出Vera Rubi…
模型 / Agent
6/22 23:05
IT之家 6 月 23 日消息，根据 Anthropic 公司新版隐私政策，该公司可能会要求 Claude 用户上传政府签发的证件来验证其年龄和身份。 这家人工智能企业表示，推出该项举措，是为了给账号被标记为存在潜在欺诈行为的用户提供申诉渠道，而非直接封禁账号。与此同时，Anthropic 正试图缓和与特朗普政府之间的矛盾，双方围绕谁有权使用该公司 AI 工…
AI 点评 · AI安全与合规升级，平衡用户隐私与反欺诈需求，折射行业监管趋严趋势。
论文 / 方法
6/22 20:00
Generalist value models play a pivotal role in scaling robotic policy learning from large-scale, mixed-quality data. Mathematically, accurate value estimation demands deep temporal understanding, requ…
论文 / 方法
6/22 20:00
Jailbreak attacks reveal a persistent weakness in aligned Large Language Models: carefully crafted prompts can elicit policy-violating responses despite safety training. While most defenses operate at…
论文 / 方法
6/22 20:00
On-policy distillation (OPD) trains a student on its own rollouts guided by teacher feedback and is becoming increasingly important for large language model (LLM) post-training. Like reinforcement lea…
行业信号
6/21 23:17
36氪获悉，中信建投研报称，市场短期呈现强劲反弹行情，但反弹完成后A股或将进入震荡期。继续看好AI算力作为本轮行情中长期核心主线的后续表现。陆家嘴金融论坛等国内政策红利有望继续推动双创板块上涨，预计创业板和科创板将继续保持强势。而估值高企和交易结构可能成为后续科技主线阶段性切换的主要原因。重点关注AI算力、工业金属、石化、机械、新能源等，逢低布局：银行、非银…
AI 点评 · AI算力获机构力挺，中长期主线地位明确，政策红利与板块轮动成焦点。
行业信号
6/21 23:13
英特尔将为苹果代工芯片 库克称 iPhone 涨价不可避免 美国科技公司政策转向，限制员工 AI 成本 Modos 推出开源 13.3 寸彩色墨水屏 Manus 原投资人计划将其原价购回 CMF 已取消今年发布计划 看看就行的简讯 少数派的近期动态 你可能错过的好文章 查看全文
论文 / 方法
6/21 20:00
On-policy distillation (OPD) improves LLM reasoning by training a student model on its own generated outputs, but standard OPD treats all student-generated outputs (SGOs) equally regardless of their i…
论文 / 方法
6/21 20:00
Vision-language models (VLMs) are increasingly deployed in consumer, medical, financial, and enterprise applications. This broad deployment expands the safety surface: risks can arise from multimodal…
论文 / 方法
6/20 20:00
Vision-Language-Action (VLA) models provide a unified paradigm for robotic manipulation, yet their real-world deployment is often bottlenecked by execution efficiency. While existing efforts predomina…
行业信号
6/20 09:30
挪威首相斯特勒6月19日表示，为防止对学习产生负面影响，挪威将禁止小学生使用生成式人工智能工具，同时限制人工智能工具在高年级学生教育中的使用。根据挪威政府公布的方案，6到13岁的小学生原则上不得使用人工智能工具；14到16岁的初中生可在教师的严密监管下谨慎使用此类工具。17到19岁的高中生应学习如何恰当地使用人工智能工具，以便为后续的高等教育和未来的职场环境…
AI 点评 · 全球首个国家层面禁止小学生用AI，反映教育界对技术冲击的深层忧虑。
行业信号
6/19 14:48
IT之家 6 月 19 日消息，据美联社报道，当地时间周四，美国联邦能源监管委员会下令，要求 6 家区域电网运营商加快 大型用电设施接入美国老化且效率低下的输电系统 。委员会表示，AI 数据中心耗电量巨大，用电需求又在迅速增长，美国必须加快并网流程。 美国能源部长克里斯 · 赖特此前敦促委员会采取行动，希望增强美国在 AI 领域与中国竞争的能力。 科技公司和…
AI 点评 · AI电力需求倒逼电网改革，揭示AI发展对基础设施的深层依赖。
论文 / 方法
6/18 20:00
Latent action pretraining learns representations of visual change from pairs of observations, but existing methods typically encode each transition as a single unstructured representation that entangl…
行业信号
6/18 19:59
OpenAI is bulking up before its IPO, landing Transformer co-inventor Noam Shazeer from Google DeepMind and former Trump AI policy official Dean Ball in the same week.
AI 点评 · OpenAI上市前挖角Transformer发明者与政策要员，人才储备信号强烈。
论文 / 方法
6/18 17:25
Prior work has shown that in-context demonstrations can jailbreak language models, but it remains unclear how models interpret different types of compliance demonstrations. We study this by mixing ben…
论文 / 方法
6/17 20:00
Policy-adherent tool-calling agents in customer-service domains must maintain task states across turns while calling tools and obeying domain policies. Task states consist of relevant facts, identifie…
论文 / 方法
6/17 20:00
Achieving dexterous robotic manipulation in the real world heavily relies on human supervision and algorithm engineering, which becomes a central bottleneck in the pursuit of general physical intellig…
论文 / 方法
6/17 20:00
MLLM-based mobile GUI agents have made substantial progress in UI understanding and action execution, but adapting them to real target apps remains costly because mobile apps are numerous, frequently…
行业信号
6/17 16:27
Hello and welcome to Regulator, an email for Verge subscribers about technology, politics, and what happens when science crashes headlong into self-interest. Not a subscriber? Sign…
行业信号
6/17 12:08
大公司： 监管约谈山姆后，沃尔玛中国多位高管变动 36氪了解到，山姆会员商店中国 CMO（首席采购官）张青已于近日提交辞呈。此外，据36氪独家了解，沃尔玛中国本次人事变动还包括：Tony Paladinetti，现任沃尔玛国际部战略副总裁，将加入沃尔玛中国，担任沃尔玛中国战略副总裁，直接向朱晓静汇报。内部邮件称，Tony将在完成工作签证流程后于2026年8月…
论文 / 方法
6/16 20:00
On-policy self-distillation (OPSD) trains a model on its own rollouts and uses a frozen copy to provide dense token-level targets conditioned on a reference target. This works well for LLM reasoning,…
论文 / 方法
6/16 20:00
Reinforcement Learning with Verifiable Rewards algorithms like GRPO have emerged as the dominant post-training paradigm for complex reasoning in LLMs, yet commonly suffer from policy entropy collapse…
论文 / 方法
6/16 20:00
Current agentic robot systems can write executable Code-as-Policy programs, observe feedback, and revise behavior across multiple attempts, but they remain largely task-driven: reusable skills are acq…
论文 / 方法
6/16 17:59
Robots deployed in the real world should learn from their experience and improve over time. This requires a mechanism of practicing and learning from feedback. In this paper, we propose VERITAS, a gen…
论文 / 方法
6/16 17:50
With sophisticated cyber-attacks becoming increasingly prevalent, modern networks require intelligent autonomous cyber-defense agents trained via Reinforcement Learning (RL). These agents employ neuro…
行业信号
6/15 23:56
今日热点导览 国内品牌金饰价格普涨 小红书公布世界杯开赛数据 阿里巴巴达摩院退出旗下科技公司 罗马仕接连变更法定代表人 遇见小面创始人向“渝见小面”致歉 TOP3大新闻 针对多发食品安全问题，市场监管总局约谈山姆总部 记者15日了解到，针对一段时期以来监管发现和媒体曝光的山姆线下门店及线上网店多发的食品安全问题，近日，市场监管总局依法对沃尔玛（中国）投资有限…
AI 点评 · 山姆被约谈、智谱股价飙升、美伊达成协议，三件大事同日凸显监管、科技与地缘政治的联动影响。
论文 / 方法
6/15 20:00
On-policy self-distillation (OPSD) has proven effective for post-training large language models (LLMs), yet its application to diffusion LLMs (dLLMs) remains unexplored. Existing OPSD methods are inhe…
论文 / 方法
6/15 20:00
Knowledge distillation transfers a teacher's competence to a small student but is brittle in the small-student regime: forcing the student to imitate logits from a much larger teacher concentrates it…
论文 / 方法
6/15 20:00
Memory has become a standard substrate for self-evolving agents, yet retaining experience is not the same as learning how to evolve through it. Existing memory agents can store trajectories, retrieve…
论文 / 方法
6/15 20:00
Graphical user interface (GUI) grounding requires vision-language models (VLMs) to identify small target elements in high-resolution screenshots and predict precise screen coordinates. On-policy self-…
论文 / 方法
6/15 17:44
Heterogeneous Treatment Effect (HTE) identification is crucial to explain the impact of an intervention and optimize our policies accordingly. Existing approaches trade expressivity for interpretabili…
行业信号
6/15 17:44
For months, Big Tech's Washington lobbyists have chased after the holy grail of pro-AI legislation: preemption. This would be a comprehensive federal law, passed in Congress and si…
论文 / 方法
6/14 20:00
As LLMs advance, post-training reinforcement learning (RL) increasingly relies on multi-dimensional rewards to cultivate comprehensive capabilities. This shift demands new algorithms capable of optimi…
论文 / 方法
6/14 20:00
Generalist robot policies must follow user instructions while reasoning about how objects, cameras, and robot actions interact in the 3D physical world. Recent vision-language-action models (VLAs) and…
论文 / 方法
6/14 20:00
Reinforcement learning with verifiable rewards (RLVR) improves language-model reasoning, but GRPO-style optimization remains prone to collapse. We analyse this instability through token-level gradient…
论文 / 方法
6/13 20:00
Extending a vision-language-action (VLA) policy to a new task typically requires task-specific teleoperated demonstrations and per-task fine-tuning, making adaptation costly in both data collection an…
行业信号
6/11 20:20
An expert in behavioral science and transportation, Zhao combines these studies with AI and public policy to address some of the most urgent challenges facing cities.
AI 点评 · 行为科学与交通专家赵华将AI融入城市政策，为解决城市紧迫挑战带来新视角。
论文 / 方法
6/11 20:00
Large reasoning models typically follow a read-then-think paradigm: they observe the complete input, reason over a static context, and then produce the answer. Yet many real-world scenarios are inhere…
论文 / 方法
6/11 20:00
When applying Group Relative Policy Optimization (GRPO) for GUI Grounding, rollouts are sampled from a single screenshot view; groups often become either all failures on difficult instances or all suc…
论文 / 方法
6/11 17:54
On-policy distillation (\textsc{OPD}) has recently become a prominent post-training recipe as it combines two desirable ingredients: on-policy student trajectories and dense teacher supervision, yet h…
Skill / 资源
6/11 03:45
Anthropic Walks Back Policy That Could Have ‘Sabotaged’ AI Researchers Using Claude Big scoop for Maxwell Zeff at Wired: “We’re changing Fable 5’s safeguards for frontier LLM devel…
行业信号
6/11 03:41
https://archive.ph/yxYhU https://web.archive.org/web/20260611033414/https://www.wired...
行业信号
6/10 23:41
36氪获悉，中信建投研报指出，6月行业配置建议采用“杠铃结构”：一端保留AI、半导体和出口制造等高景气底仓，另一端配置高股息、现金流稳定资产控制组合波动，中间用周期涨价、新能源修复和政策订单主题做弹性仓位。5月市场主线高度集中在通信、电子、AI硬件链，说明资金仍围绕产业景气和业绩兑现定价，但AI链短期涨幅较大、交易拥挤度上升，6月不宜继续提升AI总仓位，内部…
AI 点评 · 杠铃策略兼顾进攻与防御，精准捕捉6月市场结构性机会。
论文 / 方法
6/10 20:00
Latent chain-of-thought compresses reasoning by replacing visible reasoning traces with continuous hidden-state recurrence, but existing formulations are difficult to optimize with standard on-policy…
论文 / 方法
6/10 20:00
The potential impacts of world models (WMs, i.e., learned simulators) on robotics are far-reaching -- policy evaluation, policy improvement, and test-time planning -- all with limited real-world inter…
论文 / 方法
6/10 20:00
On-policy distillation (OPD) has recently become a prominent post-training recipe as it combines two desirable ingredients: on-policy student trajectories and dense teacher supervision, yet how this h…
行业信号
6/10 18:36
AI 点评 · 关注点在于政策如何应对AI指数级增长带来的安全与治理挑战。
论文 / 方法
6/10 17:59
Contact-rich manipulation requires force sensitivity, but many robot arms lack dedicated force sensors due to their high cost. We present Neural External Torque Estimation (NEXT), a data-driven method…
论文 / 方法
6/10 17:47
Recent advances in agentic Reinforcement Learning (RL) have substantially improved the multi-turn tool-use capabilities of large language model agents. However, most existing methods assign credit ove…
论文 / 方法
6/10 17:38
Human-in-the-loop reinforcement learning (HiL-RL) has emerged as an effective paradigm for real-world robotic manipulation, enabling online policy improvement with human guidance. However, current HiL…
行业信号
6/10 17:38
Hello and welcome to Regulator, a newsletter for Verge subscribers about tech politics, tech influence, and tech shenanigans in Washington, DC. (If you're not a subscriber, you can…
AI 点评 · AI监管正让立场迥异的群体被迫合作，折射出政策博弈的复杂与紧迫。
论文 / 方法
6/10 17:34
We propose Ambient Diffusion Policy, a simple and principled method for imitation learning from suboptimal data in robotics. High-quality, task-specific robot data is expensive and time-consuming to c…
论文 / 方法
6/10 13:47
Recent advances in agentic Reinforcement Learning (RL) have substantially improved the multi-turn tool-use capabilities of large language model agents. However, most existing methods assign credit ove…
论文 / 方法
6/9 18:15
Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science. Given LLMs' common…
论文 / 方法
6/9 13:16
Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models. However, rollout-intensive policy optimization is o…
模型 / Agent
6/9 00:00
Explore our ambitious, people-first industrial policy ideas for the AI era—focused on expanding opportunity, sharing prosperity, and building resilient institutions as advanced int…
Skill / 资源
6/8 23:58
Given how badly burned anyone who took Apple's 2024 WWDC Apple Intelligence announcements at face value was, I'm holding to a strict "I'll believe it when I see it" policy for ever…
AI 点评 · 苹果Siri AI若在WWDC 2026兑现承诺，将扭转此前画饼争议，值得关注其实际落地。
行业信号
6/8 23:28
36氪获悉，中信建投研报指出，本周股票成交额环比出现回落，但继续维持高位，市场交投情绪仍在较高水平。当前保险板块估值具有较高配置价值，当前股价过度反映较高基数带来的潜在业绩增速压力，资产端回暖有望带动二季度利润增速和股价修复。港股非银板块在低估值特征与盈利改善预期共振下中长期配置价值凸显。监管逻辑趋稳、促消费导向明确、AI技术提效的背景下，消费金融行业处于政…
AI 点评 · 市场活跃叠加低估值，非银及消费金融板块存修复与配置机遇。
论文 / 方法
6/8 20:00
Recent work has demonstrated that online reinforcement learning (RL) can substantially improve the quality and alignment of flow matching models for image and video generation. Methods such as Flow-GR…
论文 / 方法
6/8 20:00
The success of Large Language Models in mathematical reasoning relies heavily on the generation of diverse and valid solution paths during the rollout phase. However, current rollout techniques face a…
论文 / 方法
6/8 17:59
Training reinforcement learning (RL) policies from scratch is costly: it requires careful reward and environment design, extensive tuning, and substantial computation. Yet many control problems alread…
AI 点评 · 无需从头训练策略，巧妙迁移智能体，大幅降低强化学习成本。
论文 / 方法
6/8 17:58
Reinforcement learning (RL) has become a key component of post-training large language models (LLMs). In practice, LLM RL is often off-policy because of training-inference mismatch and policy stalenes…
AI 点评 · 反思大模型强化学习中的散度正则化，揭示当前方法缺陷，为提升训练效率提供新视角。
论文 / 方法
6/7 20:00
World-action models have emerged as a promising paradigm for robot manipulation, jointly modeling visual scene dynamics and actions to inject physical priors into policy learning. However, existing wo…
论文 / 方法
6/7 20:00
Reinforcement learning (RL) has become a key component of post-training large language models (LLMs). In practice, LLM RL is often off-policy because of training-inference mismatch and policy stalenes…
论文 / 方法
6/7 20:00
Visual reasoning requires integrating evidence distributed across regions, attributes, and relations, making single-chain reasoning prone to early perceptual commitment and hallucination. We propose V…
论文 / 方法
6/7 20:00
On-policy distillation (OPD) trains a student on its own trajectories with dense per-token supervision from a stronger teacher, and often outperforms off-policy distillation and standard reinforcement…
论文 / 方法
6/6 20:00
On-policy distillation (OPD) has become a central post-training tool for large language models (LLMs), providing dense per-token teacher supervision along the student's own rollouts. In this work, we…
行业信号
6/6 17:42
Krishnan is reportedly starting a new institution to continue shaping Trump's AI policy.
AI 点评 · 白宫AI顾问离职创办新机构，显示特朗普政府AI政策走向独立智库化。
论文 / 方法
6/5 20:00
World Action Models (WAMs) extend robot policy learning by incorporating future prediction as an additional training objective, encouraging the policy to encode task-relevant temporal structure in its…
行业信号
6/5 12:21
加拿大政府4日正式公布名为“全民人工智能”（AI for All）的国家人工智能新战略。该战略计划投入数十亿加元公共资金，通过设立技术增长基金、扩大主权算力基础设施以及加强法律监管等核心举措，加快构建本土人工智能产业体系，全面推动技术普及与主权安全。新战略的核心举措之一是设立规模达5亿加元（约合3.6亿美元）的“加拿大技术增长基金”（Canadian Tec…
论文 / 方法
6/4 20:00
On-policy distillation (OPD) is increasingly used to improve large language model reasoning, but its training dynamics remain poorly understood. We characterize the trajectory of OPD updates in parame…
论文 / 方法
6/4 20:00
Vision-Language-Action (VLA) models are emerging as a promising paradigm for robotic manipulation, enabling general-purpose policies trained from large corpora of demonstrations and action labels. How…
论文 / 方法
6/4 20:00
Agentic reinforcement learning (RL) is emerging as a critical post-training paradigm for improving LLM agent capabilities. Existing RL algorithms for LLMs largely follow the token-centric paradigm as…
论文 / 方法
6/4 17:56
Recent advancements in reasoning language models have been driven by Reinforcement Learning (RL) fine-tuning. Most often, these rely on the Group Relative Policy Optimization (GRPO) algorithm or modif…
论文 / 方法
6/4 17:50
As autonomous LLM agents increasingly hold real credentials and operate infrastructure without a human in the loop, operators have no standard way to tell an agent that a resource is off-limits. Acces…
论文 / 方法
6/3 20:00
Autonomous driving requires reasoning about how ego actions shape the evolution of the surrounding world. However, most end-to-end methods rely on direct state-to-action mappings, capturing correlatio…
论文 / 方法
6/3 20:00
On-policy distillation (OPD) supervises the student only in output space by matching next-token probabilities. This output-only paradigm has two limits: (1) sampling variance from Monte Carlo KL estim…
论文 / 方法
6/3 20:00
We study the transformation of autoregressive models (ARLMs) into diffusion language models (DLMs). Rather than pretraining from scratch, prior work replaces the causal attention in ARLMs with bidirec…
论文 / 方法
6/3 20:00
Generalist robot intelligence is often framed as a policy-scaling problem: collect more robot demonstrations, train larger Vision-Language-Action (VLA) models, and expect broader generalisation. In th…
行业信号
6/3 14:58
U.K. regulators are requiring Google offer a tool allowing website publishers to opt-out of generative AI search features. The option will be tested in the U.K. then rolled out glo…
模型 / Agent
6/3 10:00
OpenAI outlines its public policy agenda for AI, including safety, youth protection, workforce transition, and global standards to ensure AI benefits society.
行业信号
6/3 08:45
Online publishers are getting more control over whether their websites appear in Google's AI Search features, thanks to a UK regulatory ruling. The new conduct rule imposed by the…
设计 / 产品
6/3 02:20
IT之家 6 月 3 日消息，据财经杂志报道，腾讯人士表示，目前无法确定微信 AI 智能体何时推出，其上线时间很大程度上取决于监管方对智能体的审批进度，微信 14 亿的用户体量，合规流程可能比其他产品更加严格。关于微信智能体，腾讯相关负责人表示暂无回应。 IT之家注意到，此前英国《金融时报》报道称，微信将推出一款 AI（人工智能）智能体，计划最快将于本月启动…
AI 点评 · 监管审批成关键变量，14亿用户规模下的合规挑战值得关注。
设计 / 产品
6/3 02:15
昨日，媒体报道称，微信将推出一款AI（人工智能）智能体，计划最快将于本月启动公开上线前所需的合规审批流程。腾讯人士表示，目前无法确定微信AI智能体何时推出，其上线时间很大程度上取决于监管方对智能体的审批进度，微信14亿的用户体量，合规流程可能比其他产品更加严格。关于微信智能体，腾讯相关负责人表示暂无回应。（财经）
AI 点评 · 微信14亿用户体量下，AI智能体合规审批进度成焦点，决定产品上线时间。
行业信号
6/2 23:02
IT之家 6 月 3 日消息，美国总统唐纳德・特朗普于当地时间周二签署了一项行政令，推出人工智能企业自愿合作机制：AI 公司在前沿大模型正式发布前，可向美国联邦政府提交模型，以此“推动安全创新、强化关键基础设施网络安全防护”。 该行政令提出，美国人工智能产业能够蓬勃发展，部分原因在于“美方拒绝依靠繁重冗余的监管扼杀技术创新”，但同时新规也承认，新一代人工智能…
AI 点评 · 行政令体现美国平衡AI创新与安全的新思路，自愿评估机制值得关注。
论文 / 方法
6/2 20:00
Rubric-based reinforcement learning (RL) uses an LLM-as-a-Judge (LaaJ) to score model outputs according to rubrics as rewards. However, policy models may exploit latent biases in the judge, leading to…
行业信号
6/2 18:00
The specification lets developer, compliance, and security teams define their own policies for agents to follow in portable policy files.
AI 点评 · 微软推出便携式策略文件，让开发者自主定义AI代理行为规范，提升安全可控性。
论文 / 方法
6/2 17:50
Classical reinforcement learning (RL) typically seeks a deterministic policy that maximizes the expected sum of a scalar reward. Yet, modern applications such as language model fine-tuning or scientif…
论文 / 方法
6/2 17:37
We implement simultaneous translation capability with the offline direct speech-to-text translation model Canary, using the state-of-the-art policy AlignAtt, and submit it to IWSLT 2026 Simultaneous S…
论文 / 方法
6/2 13:20
Self-distilled policy optimization (SDPO) has become a popular paradigm for LLM post-training, where a model learns from its own predictions conditioned on privileged information. SDPO, however, is se…
行业信号
6/2 12:32
A new AI compliance service sits between AI models and end users to flag and replace any messages that might present a compliance problem.
AI 点评 · 融资千万美元，在AI输出端拦截违规内容，开辟模型安全新赛道。
行业信号
6/2 09:00
IT之家 6 月 2 日消息，据路透社昨天报道，欧盟计划对高度敏感的政府云项目招标提出新标准，将微软、亚马逊和谷歌等美国大型科技公司排除在外。 IT之家援引路透社，本提案属于欧盟委员会《云与人工智能发展法案》的一部分，预计将在周三正式公布。 旨在进一步减少美国企业依赖 ， 并推动欧洲本土产业发展 。 同时，欧盟不断在银行、能源和医疗等敏感领域实施“数字主权”…
AI 点评 · 欧盟加速“数字主权”战略，限制美企参与云招标，将重塑全球科技竞争格局。
论文 / 方法
6/1 20:00
As autonomous vehicle capabilities advance, the safe evaluation of driving policies in long-tail scenarios remains a critical bottleneck. In closed-loop simulation, the driving policy model actively i…
论文 / 方法
6/1 20:00
Reinforcement learning (RL) has become a dominant post-training paradigm, enabling large language models (LLMs) to learn from rewards. We observe that societal regulations are structurally similar to…
论文 / 方法
6/1 20:00
We identify a new dimension for enhancing rollout diversity in Group Relative Policy Optimization (GRPO) for LLMs. While GRPO relies on diverse rollouts, prevailing strategies primarily increase diver…
Skill / 资源
6/1 17:54
In this post, we use a lakehouse data agent to demonstrate how you can use Policy for deterministic access control and Lambda interceptors for dynamic validation. We then show how…
AI 点评 · 亚马逊Bedrock新功能实现AI代理安全管控，结合策略与动态验证，为行业提供可落地的防护方案。
论文 / 方法
6/1 17:38
Aligning Large Language Models (LLMs) with human values often degrades their general capabilities, termed the alignment tax. Existing methods mitigate this by balancing dual objectives, which heavily…
AI 点评 · 用局部策略蒸馏降低对齐成本，在保持安全性的同时避免通用能力下降，为高效安全对齐提供了新思路。
模型 / Agent
6/1 17:00
Our approach to AI policy and political advocacy, transparency, support for thoughtful regulation and AI safety, and that no outside political group speaks on the company’s behalf.
AI 点评 · OpenAI公开AI政策立场，强调透明度与安全监管，体现科技巨头在AI治理中的责任与影响力。
论文 / 方法
6/1 03:00
LLM agents are increasingly expected to operate across heterogeneous task regimes that require distinct execution paradigms. This challenges fixed agent systems and motivates system-level meta-adaptat…
论文 / 方法
5/31 20:00
Reinforcement learning (RL) improves large language model (LLM) agents by teaching them which actions lead to high rewards, but provides little supervision on what those actions do to the environment.…
论文 / 方法
5/31 20:00
On-Policy distillation (OPD) in large language models is shifting from full-trace KL supervision toward more selective training paradigms. Recent OPD methods increasingly focus on selecting which traj…
论文 / 方法
5/30 20:00
On-Policy Distillation (OPD) is a fundamental technique for efficient post-training of large language models (LLMs), with broad applications in agent learning, multi-task enhancement, and model compre…
行业信号
5/29 10:00
Pope Leo XIV’s new encyclical on artificial intelligence includes a statement that warrants serious attention from technologists and policymakers: “Technology is never neutral.” Ma…
AI 点评 · 教皇通谕点破技术非中性，为AI伦理提供了超越功利主义的道德框架，值得科技与政策界深思。
论文 / 方法
5/28 20:00
Building strong reward models (RMs) for language model alignment is bottlenecked by the cost and difficulty of acquiring diverse and reliable preference data from human annotation or judge models. It…
AI 点评 · 用模型自身生成数据改进奖励模型，突破人工标注瓶颈，是RLHF的高效自监督路径。
论文 / 方法
5/28 20:00
Group-advantage-based reinforcement learning methods, such as GRPO and DAPO, have demonstrated strong performance across diverse domains, including mathematical reasoning and text-to-image generation.…
论文 / 方法
5/28 20:00
On-policy distillation (OPD) trains a student on prefixes sampled from its own policy while matching a stronger teacher. This addresses the prefix mismatch of offline distillation, but early student r…
论文 / 方法
5/28 20:00
Video world models (WMs) have shown promise for policy evaluation and improvement by imagining realistic future observations conditioned on ego-robot actions. While WMs can model distributions over fu…
AI 点评 · 用压力测试场景驱动视频世界模型，提升机器人策略评估的鲁棒性和改进效果。
行业信号
5/28 17:01
Here’s why Anthropic and OpenAI are on board with Illinois safety testing.
AI 点评 · 伊利诺伊州新法削弱联邦对AI监管主导权，获Anthropic和OpenAI支持，凸显行业对安全测试的
模型 / Agent
5/28 00:00
Explore OpenAI’s Frontier Governance Framework and how our AI safety, security, and risk practices align with emerging EU and California regulations.
AI 点评 · 前沿治理框架首次将AI安全实践与欧盟、加州法规对标，为行业合规提供范本。
论文 / 方法
5/27 20:00
While GUI agents have advanced rapidly, they often lack the robustness to recover from their own errors, hindering real-world deployment. To bridge this gap at both the evaluation and data levels, we…
AI 点评 · 为GUI智能体提供自我纠错能力评估基准与轨迹合成方法，填补了实际部署中的关键空白。
论文 / 方法
5/27 20:00
Reinforcement learning (RL) can be used to improve the policy (denoiser) of diffusion large language models (dLLMs), while being hindered by the intractability of the policy likelihood. A dominant and…
论文 / 方法
5/27 20:00
Speculative decoding accelerates large language model inference by pairing a target model with a lightweight draft model whose proposed tokens are verified in parallel. A common way to build draft mod…
AI 点评 · 通过在线策略蒸馏提升推测解码效率，为加速大模型推理提供了更优训练方案。
论文 / 方法
5/27 20:00
When a large language model under reinforcement learning commits a wrong reasoning step early in a trajectory, standard algorithms force it to keep generating until the maximum horizon, spending compu…
AI 点评 · 用早停机制提升强化学习训练大模型效率，大幅减少无效计算资源浪费。
论文 / 方法
5/27 20:00
Memory-augmented LLM agents tackle complex long-horizon tasks by recursively summarizing interaction trajectories into compact memory. However, existing approaches typically train these memory policie…
论文 / 方法
5/25 20:00
On-policy distillation (OPD) trains a student on its own rollouts with token-level teacher supervision. Recent selective OPD methods exploit the non-uniformity of OPD signals by prioritizing high-entr…
论文 / 方法
5/25 20:00
Off-policy reinforcement learning of pretrained flow policies remains challenging due to the instability of optimization arising from the multi-step sampling process. Recently, Q-learning with Adjoint…
论文 / 方法
5/24 20:00
Reinforcement learning with verifiable rewards (RLVR) has become a core technique for post-training of Large Language Models (LLMs). While policy optimization is driven by all sampled tokens under a g…
AI 点评 · 时间调度让强化学习从空间维度扩展到时间维度，为提升推理效率开辟新思路。
论文 / 方法
5/21 20:00
Multi-agent LLM workflows route inference through specialized roles to lift end-task accuracy, but jointly training those roles with reinforcement learning is unstable in ways that are poorly understo…
AI 点评 · 多智能体强化学习提升大模型协作效率的关键在于理解分工规模与策略共享的权衡。
设计 / 产品
5/11 22:55
Runtime security monitoring and control for AI agents. Catches malicious tool use, prompt injection, and policy drift in real time, before the agent acts.
论文 / 方法
5/11 12:46
Welcome to Import AI, a newsletter about AI research. Import AI runs on arXiv, cappuccinos, and feedback from readers. If you’d like to support this, please subscribe. Subscribe no…