Bo Li

We revisit retrieval-augmented generation (RAG) by embedding retrieval control directly into generation. Instead of treating retrieval as an external intervention, we express retrieval decisions within token-level decoding, enabling end-to-end coordination without additional controllers or classi...

Instruction tuning relies on large instruction–response corpora whose quality and composition strongly affect downstream performance. We propose A nswer D ivergence- G uided Selection (ADG) , which selects instruction data based on the geometric structure of multi-sample outputs. ADG draws severa...

Instruction-tuned language models increasingly rely on large multi-turn dialogue corpora, but these datasets are often noisy and structurally inconsistent, with topic drift, repetitive chitchat, and mismatched answer formats across turns. We address this from a data selection perspective and prop...