Daily AI Papers Briefing (2026-09-04)

Podliアプリで聴く 🎧

お気に入りの番組をフォローして、オフラインでも車内でもCarPlayやAndroid Autoで再生。いつでも続きから聴けます。無料でお試しできます。

【本日の論文】
1. Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills
https://huggingface.co/papers/2609.02749
2. HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?
https://huggingface.co/papers/2609.01437
3. Aspire: Can Models Self-Evolve from Vague Goals?
https://huggingface.co/papers/2608.31111
4. SolarWM: Open Data and Scalable Training for Long-Horizon Video World Models
https://huggingface.co/papers/2609.02886
5. EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction
https://huggingface.co/papers/2609.02783