arXiv:2611.23824v1 Announce Type: new Abstract: Reinforcement learning from terminal answer rewards enables search-augmented language models to interleave reasoning and retrieval without step-level annotations, yet outcome supervision assigns identical credit to every search in a trajectory regardless of whether it retrieved useful information. This limitation is especially severe when all sampled
Straipsnis generuojamas iš faktų karkaso pasirinktu stiliumi — be išgalvojimų.
Paskelbta: 2026-11-20 05:00