Han Fang

Generalized Parallel Scaling with Interdependent Generations

See, Think, Hallucinate: Interpreting Reasoning and Hallucinations Beyond the First Hop in Vision-Language Models

Think Smarter not Harder: Adaptive Reasoning with Inference Aware Optimization

Boosting LLM Reasoning via Spontaneous Self-Correction

The Perfect Blend: Redefining RLHF with Mixture of Judges