AI 情报文章归档

Nikon 称 Dr. Ning Xu 原获 Small World in Motion 大赛第一名的视频（声称展示 PCD 患儿气道纤毛运动）不符合关于生成式 AI 的比赛规则。

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass ...

阶跃星辰宣布 Step 5 Preview 在 Kilo 限时免费一周。该模型为 600B 参数、每 token 激活 27B，支持 1M token 上下文窗口和视觉能力，主打较低成本下的编码与金融表现。

Language models increasingly support or automate information retrieval from document collections too large to index. This task is difficult when it requires specialist linguistic or scientific knowledge. We used the Wikipedia Game to test …

Interpreting machine learning model predictions is a central challenge, particularly for black-box vision models. Concept-based attribution methods interpret models using human-understandable concepts. Existing concept-based methods can on…

Hello, I'm Rijul, and I'm building LiveReview — a blast-radius aware AI code review built for your...

The expressivity of a neural network $Φ$ describes the class of functions that a network can express. The term topological expressivity refers to a family of measures particularly suited to understand binary classification and is often for…

Capability evaluations aim to surface dangerous model behaviors before deployment, but their reliability depends on the assumption that hidden capabilities can be elicited. We show this assumption does not hold under adversarial conditions…

This is a submission for the Kaggle Benchmarking Challenge. This would probably needs its own...

Anthropic 推出免费的 OSS AI 扫描器，可定期检查开源项目、自动标记并解释漏洞、给出补丁建议。预期准确率超过 90%，但报告未经人工审核可能出错；关键项目维护者可通过 GitHub 选择加入。

François Chollet 提出应把科学建模为一个递归自我改进的智能系统，作为理解 AI RSI 的参照。

A parallel sampler can choose at random which coordinates to generate together. For a target $p$ and the output law $q_U$ of a schedule $U$ drawn afresh for each output, we compare the schedule-averaged divergence $\mathbb{E}_U D_{\mathrm{…

Deep-space communications operate under intermittent connectivity, long propagation delays, and relay storage constraints. In such networks, the Bundle Protocol enables store-carry-forward communication, requiring relay nodes to manage lim…

Clinical sleep medicine repeatedly documents profound divergences between objective sleep duration and subjective daytime impairment, notably exemplified by sleep state misperception in paradoxical insomnia and boundary instability in narc…

A central challenge in de novo protein design is generating proteins with desired structural and functional properties. Existing design models typically require a dedicated input representation for each property, limiting the design object…

Qwen3.8-27B, 600 steps, 12 reward channels: CISPO, REPO-R and prompt filtering lifted holdout from −0.47 to +1.80, and an advantage floor silently zeroed 96% of the signal.

Recursive self-improvement (RSI) aims to transform interaction experience into persistent improvements, yet external feedback is costly and world-model predictions may become unreliable as policies evolve. We introduce RSI-Anything, a doma…

Vision-Language Models (VLMs) increasingly analyze and reason about images, yet it remains unclear whether their internal representations faithfully encode the structured relations required for such reasoning. We introduce EICS (Evaluation…

While deep learning models for tabular data have recently received significant research attention, the choice of objective functions for regression tasks remains an understudied degree of freedom. Several lines of research offer alternativ…

当今的前沿语言模型能否零样本、无指导地控制真实机械臂？ AA-Robotics v0.1 即将发布。

We study the last iterate of standard asynchronous Q-learning from a single trajectory of a finite Markov decision process. The trajectory follows a fixed behavior policy whose state--action chain is irreducible and may be periodic. For di…

Neural semantic parsers generate meaning representations token by token, yet well-formedness depends on references, scope, and graph relations that unfold during generation. Constrained decoding prevents locally invalid tokens but does not…

Large language models reuse the same attention heads across tasks, but identifying these shared heads does not explain what they do differently in each task. A natural source of inspiration is task-dependent interactions among shared brain…

Grok Bot 现在拥有自己的邮箱，可用它注册服务、代用户联系商家或与他人约时间。作者提示可以在 X 上询问自己的 bot 或 @bot 来完成设置。

A concrete example connecting slot_mapping, block_table, cache layouts, and tiled attention.

Codex 发布 v0.162.1，修复了异步问题包含多行时导致的 TUI 崩溃，保留换行与完整超链接目标；同时修复了后台服务器功能设置与 CLI 默认值不一致引发的启动失败，兼容性检查现仅适用于显式命令行功能覆盖。

As AI agents increasingly rely on external tools, accurate retrieval is essential for reliable behavior. Existing methods optimize semantic relevance but largely ignore tool availability during execution. To address this gap, we propose RA…

Grok @Bot 现在可以自行设置邮箱了

我坐副驾！ （引用推文核心要点：让 @Muse 给我附近每一家奥迪经销商打电话，找一辆 Q9 试驾。它跟真人通了话并回报结果。目前停车场里还没有 Q9，但有一个人邀请我们去参加发布活动。完全省掉了一下午的等待音乐。@alexandr_wang @AIatMeta @Audi）

Personal agents increasingly connect to services across the web and applications by using tools. As the effects of their actions accumulate and interact across services, unintended or harmful outcomes can arise when agents fulfill benign u…

Grok 官方宣布 Grok Bot 现在拥有自己的邮箱，用户可让机器人设置地址，或在回复中 @bot 使用。Bot 可用该邮箱代用户注册服务、联系商家或与他人安排时间，图片展示地址形如 your-bot@mail.grokbot.com。

A memory that reproduces every current prediction may still have forgotten how to learn from the next observation. We characterize this distinction for exact cumulative learning and attention. For smooth convex residual losses with unrestr…

Grok Bot 今日开始向用户推送，用户可通过询问 Bot 或在 X 上标记 @bot 领取收件箱。团队使用需由管理员开启，入口为 https://x.ai/bot。

Grok Bot 现在有了自己的邮箱。 Bot 可以用它注册服务、替你联系商家，或与某人安排时间。

This is a submission for the Kaggle Benchmarking Challenge TL;DR I asked 6 vision models to...

Many optimization problems provide direct access to only part of the gradient, while the remaining component must be inferred from function evaluations. This structure arises in performative prediction, performative reinforcement learning,…

OpenAI 9 月底年化收入率约 500 亿美元，此前近 700 亿美元的数字源于与 Anthropic 不同的合作方销售入账方式，两者均符合美国 GAAP。公司正洽谈至少 300 亿美元新融资，目标投前估值 1.4 万亿美元，企业业务推动 Q3 总收入增长 77%，FT 报告发布后芯片股曾下跌数个百分点。

HeyGen Voice 在 Artificial Analysis 受控语音 TTS 竞技场以 Elo 1,201 登顶，超过 Qwen-Audio-3.1-TTS-Plus（1,182）和 Eleven v4 Turbo（1,166）。其发音鲁棒性得分 83.1%，定价 $30/1M 字符，生成速度为每秒 40 字符。

Grok Bot 宣布今日开始向用户推送，用户只需 ask Bot 即可认领自己的 inbox；团队使用需由管理员开启，入口为 https://x.ai/bot。

我收到针对 x 的钓鱼邮件时的反应