LLM Probability Concentration: How Alignment Shrinks the Generative Horizon
LLM概率集中:对齐如何缩小生成范围
专题命中 复杂问题求解 :reasoning(abstract);chain-of-thought(abstract);CoT(abstract);分类 cs.CL、cs.AI、cs.LG
AI总结 研究发现对齐微调通过减少生成多样性,使LLM生成更一致,从而影响复杂推理稳定性。
Comments Codebase: https://github.com/yangalan123/LLMBranchingFactor. V3: Significantly rewrite the whole paper for a clearer structure. Correct problems in the theory parts (Remove emphasis on AEP, discussions on variable LLM generation lengths) and strengthen asymptotic analysis. Add Qwen and OLMo2 experiments. Preliminary SFT v.s. RL comparison to better understand the alignment effects on BF