Latest

This approach shifts LLMs from universal reasoning engines to targeted semantic reasoners, handling deterministic computations exactly and efficiently. The largest gains occurred on counting questions, showing how selective reasoning improves both accuracy and inference efficiency. Developers working on autonomous systems can implement this hybrid approach to optimize performance on specific query types.

Training on internalized skill workflows lets a 35B open-weights model beat far larger proprietary models on agentic benchmarks, shifting the edge from raw model scale to curated environment data. Builders running Claude Code or Codex harnesses can reuse the released environments and trajectories for their own SFT or RL pipelines. The results give Qwen-based deployments a concrete benchmark point against Claude Sonnet 4.6 and DeepSeek V4 Pro.

The key finding is that skills can be internalized as model weights rather than injected at inference time, since the fine-tuned agent beats skill-assisted base models even without the skills provided. Agent engineers and fine-tuning practitioners can reuse the released environments and trajectories for SFT or RL with outcome-based rewards. The result also shows a 35B open Qwen-based model outscoring much larger closed models on SkillsBench, a useful cost comparison for teams building coding and tool-use agents.

For researchers in recommender systems, AutoRecLab automates the process of turning experimental designs into executable code, reducing manual work and potential errors.

ATIBA automates tedious research paper verification, saving academics time while ensuring compliance with venue-specific standards.

新基准用《秘密希特勒》测AI说谎，16个模型里GPT-5.4、Kimi K2.5领先。

这篇论文揭露了GPT-5.4和Claude在资格评估规则中会莫名崩溃，还给了个能稳定执行的神经符号方案，搞合规或应用LLM的得看看。

这篇论文展示了如何在华为Ascend芯片上高效后训练万亿参数模型，并且通过运筹学专用微调，在特定任务上超越了GPT-5.4-Mini。

这篇论文用STOCKTAKE基准把LLM智能体的‘没看懂’和‘看懂了却没行动’分开了，四个主流模型的表现差距比想象中大。

这篇研究告诉你，红外视觉语言模型有多脆弱——一个 QR 风格的边角补丁就能让 OpenAI CLIP 准确率从 98.67% 跌到 0.70%，还能黑盒攻击其它模型。搞安全或对抗训练的值得一看。

这篇论文用实验告诉你，临床AI回答EHR问题时，推理步骤越多越容易翻车。Claude、GPT-4o和GPT-5.4都逃不过，部署前得重点防多步推理。

做 AI 编程智能体或评估基准的团队，这篇论文揭示了主流基准（如 SWE-Bench）掩盖的能力差距——强智能体在陌生语言上的元编程策略值得借鉴，建议点开看具体实现方法。

做推荐系统或对话智能体评估的团队，终于有了一个可复现、低成本的客观基准，直接拿来测自己的模型会看到真实差距。

做特殊教育NLP或低资源语言模型微调的团队，这篇论文提供了一个可复现的CGFD流程，直接解决了繁体中文IEP生成的数据稀缺和隐私问题，值得点开看具体实现。

AI 安全研究者终于有了评估隐性操纵的专用工具——CogManip 覆盖 15 种策略、1000 个场景，做模型对齐和红队测试的团队可以直接拿来用。

该研究揭示了记忆增强AI代理在敏感信息使用上的关键盲区，做对话系统和个性化AI的开发者值得关注——它直接关系到用户隐私和信任。