研究

<a href="https:&#x2F;&#x2F;deploymentsafety.openai.com&#x2F;gpt-5-6&#x2F;gpt-5-6.pdf" rel="nofollow">https:&#x2F;&#x2F;deploymentsafety.openai.com&#x2F;gpt-5-6&#x2F;gpt-5-6.pdf</a><p><a href="https:&#x2F;&#x2F;developers.openai.com&#x2F;api&#x2F;docs&#x2F;guides&#x2F;latest-model" rel="nofollow">https:&#x2F;&#x2F;developers.openai.com&#x2F;api&#x2F;docs&#x2F;guides&#x2F;latest-model</a><p><a href="https:&#x2F;&#x2F;x.com&#x2F;levie&#x2F;status&#x2F;2075287443411222628" rel="nofollow">https:&#x2F

💬 评论区普遍认为GPT-5.6在基准测试和成本效率上超越Claude Fable，但也有人质疑其编码能力仍落后于Opus，且担忧定价过高。

<p>OpenAI's latest flagship model <a href="https://openai.com/index/gpt-5-6/">hit general availability this morning</a>, and comes in three sizes: Luna, Terra, and Sol (from smallest to largest).</p> <p>The new models are priced per 1M input/output tokens as Luna $1/$6, Terra $2.50/$15, Sol $5/$30. For comparison, the Claude Opus series are $5/$25 and the Claude Fable 5 is $10/$50, but price-per-million tokens doesn't tell us much now that the number of reasoning tokens can differ so much betwee

OpenAI's latest family of models promises improvements across a range of areas, including cybersecurity.

The AI firm Anthropic has developed a technique that has given it the clearest glimpse yet at what’s really going on inside large language models as they answer questions or carry out tasks. What they found ranges from the mundane to the unnerving. Researchers at the company built a tool called the Jacobian lens (or&#8230;

Tomesphere Atlas 映射 850 万篇研究论文为交互式图谱，方便按领域、引用和时间浏览。

<a href="https:&#x2F;&#x2F;ai.meta.com&#x2F;static-resource&#x2F;muse-spark-1-1-evaluation-report" rel="nofollow">https:&#x2F;&#x2F;ai.meta.com&#x2F;static-resource&#x2F;muse-spark-1-1-evaluatio...</a> [pdf]<p><a href="https:&#x2F;&#x2F;developer.meta.com&#x2F;ai&#x2F;resources&#x2F;blog&#x2F;build-with-muse-spark&#x2F;" rel="nofollow">https:&#x2F;&#x2F;developer.meta.com&#x2F;ai&#x2F;resources&#x2F;blog&#x2F;build-with-muse...</a><p><a href="https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;ar

“已经出现泛化涌现”

The AI ROI debate has returned and the numbers are even bigger, as are, perhaps, the consequences.

I don't know where this is headed, but I don't like it. https://futurism.com/artificial-intelligence/open-source-ai-model-scary-mythos GLM-5.2 can be downloaded by anybody, can be run on virtually any hardware, and unlike Mythos or Fable, there’s no vendor playing the middle man between the AI models and the users, raising the cybersecurity stakes considerably. Put simply, while these frontier models can aid researchers in patching holes in commonly used software, the can also be abused by hacke

arXiv:2607.06720v1 Announce Type: new Abstract: Training large language models (LLMs) with extended reasoning has enabled in-context search, in which models iteratively generate, critique, and revise solution attempts. We provide a theoretical analysis of in-context search by modeling it as approximate inference over reasoning traces, where the base model defines a prior and self-reflection provides feedback for posterior updates, and study the resulting inference-time sampling complexity - the

arXiv:2607.06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time compute over frontier models (evolutionary search, exhaustive sampling, extended chain-of-thought), or benchmark-specific training in which small models are fine-tuned on ARC data, often with task-specialized architectures. We study a third regime: an open-weight model in non-thinking mode (DeepSeek V3.2) under a strict budget, with no ARC-

I am curious about intended sizings of the main size niches of the popular local LLM models. As in, we can see there is a major niche at 26b-35b, then hardly anything from 36 through 69b, then (formerly) another major niche at ~70b-72b, then another niche at ~120b-123b, then another big gap till ~230b-235b, and then it gets a bit more mixed all over the place after that with 300b-750b being scattered more randomly probably based more on just whatever the best strength per size they could get whe

arXiv:2607.07021v1 Announce Type: new Abstract: Humans continuously coordinate with others in dynamic interactions, often through implicit, hard-to-quantify social norms that act as shared tacit expectations among interacting agents. As AI agents, including large language models (LLMs), become embedded in daily life, they increasingly participate in such interactions and reshape social interaction structures. Yet they often fail to coordinate with humans in an effective, considerate, and natura

<article><div class="page-wrapper PostDetail-module-scss-module__UQuRMa__hero"><div class="PostDetail-module-scss-module__UQuRMa__illustrationHeroWrapper"><img alt="Inviting hard questions" class="PostDetail-module-scss-module__UQuRMa__heroImage" height="1125" src="https://www-cdn.anthropic.com/images/4zrzovbb/website/3fe58a56e696628496e95871e00b8287035ea645-2000x1125.jpg" width="2000" /></div></div><div class="page-wrapper"><article><div class=""><div class="Body-module-scss-module__z40yvW__bod

arXiv:2607.05804v1 Announce Type: new Abstract: 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 application to long-horizon agentic tasks remains insufficiently explored. We identify two key inefficiencies in vanilla agent OPD: (1) full-horizon rollouts often waste wall-clock resources on tail turns that provide weak and noisy KL supervision, and (2) tr

arXiv:2607.05750v1 Announce Type: new Abstract: Computer-aided design (CAD) for industrial components requires long-horizon procedural modeling, robust feature dependencies, editable parametric geometry, and production-grade B-Rep execution. Existing text-to-CAD methods have made promising progress in generating CAD programs from natural-language descriptions, but they still struggle when user prompts are ambiguous, underspecified, or only describe high-level design intent. They also rarely exp

Akashic提出MemAttention机制，通过分块记忆和语义建模降低LLM推理开销，适用于长上下文Agent场景。

FirstResearch框架为LLM科学发现Agent生成可审计的研究问题证书，记录假设与反证。

arXiv:2607.05690v1 Announce Type: new Abstract: Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store queried at most once per turn. We study the regime where memory moves inside the loop, read and written on every step. The obstacle has always been latency: networked stores answer in tens to hundreds of milliseconds, and in-loop retrieval can inflate end-to-end latency by up to 83x when retrieval is expensive. Prior work manages that cost

AgenticAI-Supervisor提供API/UI驱动的RL Gym环境，用于评估和训练LLM Agent的多步决策能力。

arXiv:2607.05775v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly evaluated on their ability to use tools, plan multi-step tasks, coordinate with other agents, and operate over extended horizons. Reported benchmark gains often obscure recurring failure modes documented across otherwise unrelated evaluation efforts. This paper synthesizes 27 benchmark, taxonomy, and audit papers (2023-2026), spanning 19 distinct benchmarks, into a cross-cutting taxonomy of agent

arXiv:2607.05571v1 Announce Type: new Abstract: Large language models are increasingly explored as AI tutors, yet deploying them in K-12 settings raises concerns around privacy, cost, and reliance on proprietary models. Small language models (SLMs) offer a promising alternative, but selecting the right model for a specific educational context remains difficult, particularly when the target domain, such as block-based programming, is largely absent from model training data. We introduce CSTutorB

arXiv:2607.05790v1 Announce Type: new Abstract: Tool-augmented large language models extend their capabilities beyond parametric knowledge through external tools, but tend to invoke them unnecessarily. We investigate whether tool-use decisions have any stable internal representation that can be extracted and manipulated, a question that is non-trivial given that tools exist entirely in context at inference time and have no direct encoding in model weights. We show that steering vectors extracte

NapMem将长程用户记忆组织为结构化动作空间，让Agent主动导航记忆而非被动检索。

SwarmResearch：通过 Shepherd Agent 协调搜索 Agent 群体进行开放式问题发现，避免单 agent 收敛问题。

Object-Centric Environment Modeling：为 LLM agent 构建可执行的对象中心环境模型，支持技能复用和持续学习。

arXiv:2607.02686v1 Announce Type: new Abstract: Reinforcement learning agents operating under partial observability must act on incomplete information, making them natural candidates for guidance from small language models (SLMs) that carry broad reasoning priors. Yet integrating SLM guidance into this setting has proven difficult: across all test environments, vanilla uncertainty-gated approaches achieve an overwrite rate at or near zero, meaning the SLM almost never contributes an independent

arXiv:2607.02914v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated remarkable capabilities across diverse applications, yet ensuring their simultaneous safety, helpfulness, and trustworthiness remains a persistent challenge. Conventional refusal-oriented alignment strategies mitigate harmful content generation but systematically fail to serve legitimate user needs, often withholding information that could safely and constructively address the underlying intent of sen

用强化学习引导 LLM 进行主动证据获取的诊断推理，在医疗场景中提升证据采集的准确性。