Xin Xu

Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effective and Efficient Reasoners

Perception-R1: Advancing Multimodal Reasoning Capabilities of MLLMs via Visual Perception Reward

MathFimer: Enhancing Mathematical Reasoning by Expanding Reasoning Steps through Fill-in-the-Middle Task

On Predictability of Reinforcement Learning Dynamics for Large Language Models

VerifyBench: Benchmarking Reference-based Reward Systems for Large Language Models

Double-Checker: Enhancing Reasoning of Slow-Thinking LLMs via Self-Critical Fine-Tuning

Teaching LLMs According to Their Aptitude: Adaptive Switching Between CoT and TIR for Mathematical Problem Solving

GPAS: Accelerating Convergence of LLM Pretraining via Gradient-Preserving Activation Scaling

UGPhysics: A Comprehensive Benchmark for Undergraduate Physics Reasoning with Large Language Models

UGMathBench: A Diverse and Dynamic Benchmark for Undergraduate-Level Mathematical Reasoning with Large Language Models

Double-Checker: Enhancing Reasoning of Slow-Thinking LLMs via Self-Critical Fine-Tuning

Can LLMs Solve Longer Math Word Problems Better?