Dayiheng Liu

ReForm: Reflective Autoformalization with Prospective Bounded Sequence Optimization

RefineX: Learning to Refine Pre-training Data at Scale from Expert-Guided Programs

JURY-RL: Votes Propose, Proofs Dispose for Label-Free RLVR

Winning the Pruning Gamble: A Unified Approach to Joint Sample and Token Pruning for Efficient Supervised Fine-Tuning

WorldPM: Understanding Scaling Patterns in Human Preference Modeling

Rethinking LLM Evaluation: Can We Evaluate LLMs with 200× Less Data?

EvolArena: An Evolving Arena for Multi-Turn Reasoning in LLMs

CALM Before the STORM: Unlocking Native Reasoning for Optimization Modeling

Unified Data Selection for LLM Reasoning

Controllable Test-Time Scaling via Sparse Autoencoder‑Based Reasoning Steering

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

Parallel Scaling Law for Language Models

Chain of Execution Supervision Promotes General Reasoning in Large Language Models

Teaching Language Models to Reason with Tools

Self-Evolving Critique Abilities in Large Language Models

DataMan: Data Manager for Pre-training Large Language Models

Language Models can Self-Lengthen to Generate Long Texts