人人能懂的AI前沿解读(11.7)

[LG] Muir: Linear Attention with Sensitivity-Allocated Gating and Matched-Filter Readout
[Google & University of Texas at Austin]
https://arxiv.org/abs/2611.08376
---
[LG] Test-Time Self-Play: Scaling without a Verifier
[ETH Zurich]
https://arxiv.org/abs/2611.07367
---
[RO] VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding
[NVIDIA]
https://arxiv.org/abs/2611.08449
---
[RO] Social Humanoid: Socially Responsive Human-Humanoid Interaction
[UC Berkeley & Sony Group Corporation]
https://arxiv.org/abs/2611.08407
---
[LG] Optimistic Hedge in General-Sum Games: Fundamental Limits and Regret Bounds via Temporal Approximation
[Tel Aviv University]
https://arxiv.org/abs/2611.07078