A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Unknown
SGD-KV compresses LLM KV caches via head-level budget allocation and token-level selection guided by summarization, building on SnapKV-style cumulative attention.
Unknown
SelecTKD introduces selective token-weighted knowledge distillation for autoregressive LLMs, formalizing token-level divergences and weighting tokens by importance.
Unknown
This paper introduces Token-level Direct Preference Optimization (TDPO), a method that extends DPO to optimize at the token level for improved alignment of LLMs with human preferences.