在Claude Code和Codex对话中嵌入书签语录，给编程注入文学灵感。
This is a small quality of life project i built for myself and using it for 3+ months now, decided to fork it out and publish as a skill separately. I…
用自训练路由代理，低成本享受前沿模型输出质量，零延迟。
Article URL: https://github.com/numinous-technology/open-ultra Comments URL: https://news.ycombinator.com/item?id=48993540 Points: 2 # Comments: 0
用自蒸馏技术增强基于评分标准的强化学习，提升训练效果和泛化能力。
arXiv:2607.18082v1 Announce Type: cross Abstract: Rubric-based RL has recently shown promise in improving LLMs on open-ended tasks. A widely recognize…
不换模型，只改Harness就能让AI性能飙升104%？上海AI Lab自我进化新方案，效率惊人！
Harness本身也可以被搜索、验证和迭代
前Snapchat工程师打造AI工具，一键将相册照片生成社交短视频，极大节省剪辑时间
The app is designed for people who want to create social content, but find traditional video editing tools too complex or time-consuming.
提出Self-GC机制，让LLM智能体在长周期任务中自主管理上下文，突破记忆与推理瓶颈。
arXiv:2607.00692v1 Announce Type: new Abstract: Long-horizon LLM agents accumulate tool results, files, plans, and user constraints that are too struc…
揭示语言模型自我生成QA训练的隐藏脆弱性，一篇值得关注的AI研究论文。
arXiv:2606.32002v1 Announce Type: new Abstract: Language models are increasingly taught from synthetic question--answer (QA) supervision: a model gene…
公平感知线性老虎机中Nash Welfare算法的优化突破，值得关注！
arXiv:2601.22969v2 Announce Type: replace Abstract: Nash regret has recently emerged as a principled fairness-aware performance metric for stochastic …
新框架让AI智能体能自行改写运行规则，性能最高提升60%，挑战传统人机协作边界
Not every company can or should build their own frontier AI language model. However, the harness controlling the model is something that most enterpri…
多智能体AI协调框架，支持自托管，让你完全掌控数据与协作过程。
Comments URL: https://news.ycombinator.com/item?id=48586547 Points: 1 # Comments: 0
针对强化学习训练中 rollout 生成缓慢的瓶颈，提出了一种系统感知的自推测解码方法，显著加速 LLM 推理。
arXiv:2606.18967v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a representative post-training paradigm for LLMs, enabling stro…
多智能体系统自动进化元技能，让AI协作更智能更高效。
arXiv:2606.18837v1 Announce Type: cross Abstract: Large Language Model (LLM)-based automatic Multi-Agent Systems (MAS) generation has become a crucial…
用Zig编写的轻量级ELF/PE二进制检查工具，是安全分析和逆向工程的好帮手。
Article URL: https://github.com/hrasityilmaz/Biber Comments URL: https://news.ycombinator.com/item?id=48458018 Points: 1 # Comments: 0
结构化API建议让AI代理错误恢复效率提升36-40个百分点，超越冗长诊断方法
arXiv:2606.05037v1 Announce Type: cross Abstract: When an AI agent calls an API and hits a validation error, it needs more than what went wrong -- it …
在线体验Zig语言链接器优化成果，编译时间从36秒降至244毫秒，极速反馈助力开发迭代
Article URL: https://ziglang.org/devlog/2026/#2026-05-30 Comments URL: https://news.ycombinator.com/item?id=48338673 Points: 221 # Comments: 96
一款将网页和LLM模型编码进URL的创意工具，支持在浏览器本地调用Claude、OpenAI等多模型，无需后端。
Article URL: https://oxedom.github.io/seed/ Comments URL: https://news.ycombinator.com/item?id=48344326 Points: 2 # Comments: 1
自托管个人AI助手Frona发布v2026.5.5版本，带来新功能与优化
Article URL: https://github.com/fronalabs/frona/releases/tag/v2026.5.5 Comments URL: https://news.ycombinator.com/item?id=48340093 Points: 1 # Comment…
自监督验证机制让大模型推测解码无需额外训练，实现2-3倍推理速度提升，架构简洁高效。
arXiv:2510.02329v2 Announce Type: replace-cross Abstract: Speculative decoding accelerates LLM inference by verifying candidate tokens from a draft mo…
揭示自我对弈仅在自合成数据提供可学习信息增益时才有效演化，为AI训练策略提供关键理论指导。
arXiv:2603.02218v2 Announce Type: replace Abstract: Large language models (LLMs) make it plausible to build systems that improve through self-evolving…
提出Krause Attention机制，解决Transformer中全局softmax导致的同步动态与表示崩塌问题
arXiv:2602.11534v3 Announce Type: replace-cross Abstract: Self-attention in Transformers relies on globally normalized softmax weights, causing all to…