MMODELYST
People/Xiang Gao
PPL

Xiang Gao

Papers (3)
DialoGPT: Large-Scale Generative Pre-training for Conversational Response GenerationIt's Not Always Sycophancy: Measuring LLM Conformity as a Function of Epistemic UncertaintyGoal-Conditioned Supervised Learning for LLM Fine-Tuning