AI 论文与研究

We introduce VideoMSN, a Masked Siamese Network framework for efficient self-supervised spatio-temporal representation learning in videos. Instead of relying on heavy 3D architectures or reconstruction-based autoencoders for learning with …

Evolutionary search with large language models (LLMs) can stall when progress requires external knowledge the model lacks. Supplying relevant documents helps, but simply adding web search tool can keep returning the same pages as solutions…

Differentially Private Stochastic Gradient Descent (DP-SGD) is a leading approach for privacy-preserving fine-tuning of large language models (LLMs). Many decoder-only LLMs employ weight tying between input and output embeddings, a design …

Automating the search for effective harnesses is an important step toward enabling agents to recursively self-improve. Existing harness optimizations typically produce a single global harness that is applied uniformly across task instances…

As interactive 3D worlds are increasingly used to study intelligent behavior, it becomes important to develop efficient pipelines for identifying anomalies in these simulated environments, such as floating objects, traversable walls, or ob…

We present Cogentic, a multi-agent harness for automated proof discovery on open research problems. While frontier language models can generate strong mathematical ideas in a single shot, single-shot generation is often insufficient for op…

Material generation should produce not only an appearance, but also the rules that construct it. We introduce MatLoom, a compact, layer-oriented language for text-to-material generation with pretrained language models. Each program compose…

Looped transformers and Mixture-of-Experts (MoE) offer complementary routes to efficient scaling: recurrence increases computational depth at fixed parameters, while MoE sparsity expands total capacity at fixed active compute. Yet existing…

Lossy compression is widely used in Federated Learning (FL) but is generally treated as an error source, while conventional poisoning defenses inspect update geometry. In this work, we instead treat the compressor's response as a security …

Pretrained robot policies provide useful action priors, but long-horizon manipulation still requires coordination between semantic reasoning and physical execution. Semantic reasoning operates at a coarser timescale than physical interacti…

Improving text-to-image models has traditionally relied on increasing model size or the number of denoising steps. In this work, we explore an alternative way to scale computation by repeatedly running shared Transformer blocks within each…

Recent autonomous machine learning engineering (MLE) agents have made significant progress on public leaderboards. Often motivated by progress stagnation over long-horizon cycles and limited Large Language Model (LLM) primitives, modern ML…

Web text makes up the majority of pretraining data and is increasingly AI-generated. After applying FineWeb quality filtering, we find that 27.5% of tokens from June 2026 web data are labeled as AI-generated by Pangram, rising to 31.1% by …

How is computation organized and reused across tasks and time in a trained recurrent network? Most analyses emphasize the geometry of neural activity, dynamical motifs, or local perturbation growth. We instead study the network's global fi…

In algorithmic statistics a string x is explained by a finite set containing it, and Kolmogorov's structure function records the smallest such model at each level of complexity. Vereshchagin's strong models, those computable from the data …

Physics-constrained generative models aim to generate physical fields that match a target distribution and satisfy prescribed constraints. However, enforcing these constraints often increases sampling costs through iterative corrections or…

LOCI 是一种混合空间记忆架构，在部分 Transformer 块中保留过去观测的 key-value 缓存，其余块仅处理当前 chunk 并辅以读写受投影相机几何条件约束的循环线性注意力记忆，使视角同时进入记忆寻址与存储内容。

发表于斯坦福 Journal of Online Trust and Safety 的研究识别了88个活跃托管未经同意 AI 生成亲密图像的网站，指出 Cloudflare、Google、Namecheap、WordPress 和 Proton 五家为主要基础设施服务商，其中 Cloudflare 提供了大部分托管、CDN 和域名解析服务，尽管各家服务条款均禁止非法内容。

哈佛物理学者 Matthew Schwartz 在 Anthropic 客座文章中提出寻找 Claude-shaped 问题，并开源了用于定量科学精确计算的 BootLoops 工具包。

微软研究院开发了一套端到端机器学习流水线，结合太阳风观测、AE 与 Dst 指数预报、地质电导率和电网数据，为美国本土 66,935 座变电站生成 30-60 分钟前的局地空间天气风险估计。

Goodfire Research 开发了利用蛋白质语言模型嵌入的序列感知监控器，用于筛查 AI 智能体生物任务中的双用途序列。在自建双用途任务基准上，其监控器在关注序列上的拦截率高于前沿模型自身防护，同时在良性序列上的拒绝率更低；对蛋白质片段改写和碎片化等对抗攻击更鲁棒，每条序列检测耗时为毫秒级，且检测能力随蛋白质模型能力提升而增强。

Google DeepMind 团队发表论文，提出基于 SynthID 的 SynthIDBio 技术，可在流行蛋白设计工具 ProteinMPNN 生成序列时嵌入不影响功能的水印，帮助 DNA 合成方筛查未知蛋白来源。

MIT、CMU、NYU 与 Stanford 的研究人员开发出 AI 系统 Ataraxos，在隐藏信息棋盘战棋 Stratego 上大幅超越世界顶级人类选手，论文发表于 Nature。

Artificial Analysis 发布 AA-Video-T2V v2.0 和无音频版 AA-Video-T2V-Silent v2.0 文生视频基准，统一以 1080p 高码率评估，按 10 种用例、10 种能力和多种视觉风格排名，分别基于超 68,000 票（1,000+ prompts）和超 47,000 票（500 prompts）的人工偏好投票。

腾讯混元联合复旦、清华发布 ExplorationBench，用可执行规则的 Alien Worlds 评测 AI 自主探索能力，含 AlienCode（31 条隐藏规则改动、70 任务）和 AlienLogic（24 条推理规则、70 定理）两个沙盒。

当前的 harness 在 LLM 如何管理自身上下文方面设置非常僵化。我们展示了你可以让 LLM 直接编辑上下文——从而实现更好的长时程记忆、FLOP 高效适配，以及更多。 不用再害怕 compaction 了！

面壁智能联合清华 THUNLP 提出 Diffusion Reward Models（DRM），不再把人类偏好压缩成单一分数，而是学习完整奖励分布，保留分歧、不确定性与多种合理判断。

BlueQubit 团队使用 IBM 120 量子比特的 Nighthawk r2 处理器，在 36 个周期、918 个双量子比特门配置下仅用 19 秒完成 100 万次随机量子线路采样，研究人员估算 Frontier 超级计算机复现同等任务约需 110 年，约 1.2 × 10²⁷ 次计算操作。

Time series anomaly detection (TSAD) is increasingly deployed in streaming settings, where data arrive sequentially and may exhibit non-stationarity. As a result, several works from the recent literature propose streaming anomaly detection…

We study additive regression under a potentially non-product random design on $[0,1]^d$, allowing the dimension $d$ to grow with the sample size $n$. We introduce coupled smoothness classes that separately control the regularity of the mar…

SpatialClaw 重新思考智能体空间推理的动作接口 论文：https://huggingface.co/papers/2606.13673

Large Audio Language Models (LALMs) rely on effective audio encoders for multi-task performance. We introduce UniAE-MoE, a unified audio encoder designed to model cross-domain audio representations and achieve outstanding downstream unders…

Wireless split learning (SL) reduces on-device computation by offloading upper layers to a server, yet transmitting high-dimensional intermediate features at each iteration remains a major communication bottleneck. Existing methods select …

A deep classifier is defined not only by the decision it produces, but also by the sequence of transformations through which that decision is formed. Treating this evolution as a dynamical system across layers provides a natural framework …

Trustworthy Legal AI requires systems that can answer legal questions while grounding their responses in authoritative sources. However, existing Vietnamese legal benchmarks provide limited coverage of real-world legal consultations. We in…

Developmental psychology holds that certain priors are given to infants prior to experience rather than induced from data, and that the influence of such priors is suppressed under strong, well-constrained conditions but reasserts itself u…

Mamba-style and hybrid language models compress their past into a fixed-size recurrent state that is rewritten at every generated token. Storing this state in low precision saves memory bandwidth, but every rounding error is fed back into …

Fiber orientation and compartmental microstructure are central to the characterization of white matter tissue in diffusion MRI, yet existing methods either resolve fiber orientations without quantifying microstructure, or quantify microstr…

World Models are appearing as the next major frontier in computer vision. However, their robustness is currently largely unexplored. We identify the phenomenon of hallucination in latent World Models: given a state and an action, the predi…

Deep neural networks tend to rely on simple features that may be spurious and thus fail to generalize. We study this problem in the setting of linear probes, where a (generalized) linear model is fitted on the representations of a (pretrai…