人人能懂的AI前沿解读(10.30)

[AI] Evolving Agent Teams
[UC Berkeley]
https://arxiv.org/abs/2610.36848
---
[LG] Action Learning for Scientific Simulations
[Kempner Institute & Microsoft Research New England]
https://arxiv.org/abs/2610.31585
---
[CL] Memory Tax: Dissecting the Net Value of Agent Memory
[University of Illinois Urbana Champaign & Tsinghua University]
https://arxiv.org/abs/2610.34073
---
[CL] On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models
[Google]
https://arxiv.org/abs/2610.32770
---
[RO] Learning Force-Reactive Humanoid Loco-Manipulation
[Amazon FAR]
https://arxiv.org/abs/2610.35862