Daily AI Papers Briefing (2026-08-20)

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. Demystifying Agent Skills: Why They Work-Until They Don't
https://huggingface.co/papers/2608.14036
2. Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements
https://huggingface.co/papers/2608.17310
3. FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution
https://huggingface.co/papers/2608.16157
4. ASI-Bench: At the Dawn of Artificial Superintelligence
https://huggingface.co/papers/2608.17271
5. Embodied-Navigator: Point, Think, Memorize, and Align for Efficient Navigation
https://huggingface.co/papers/2608.17512