Kexun Zhang

HARDTESTGEN: A High-Quality RL Verifier Generation Pipeline for LLM Algorithimic Coding

Is Vibe Coding Safe? Benchmarking Vulnerability of Agent Generated Code in Real-World Tasks

DeFT: Decoding with Flash Tree-attention for Efficient Tree-structured LLM Inference

Diversity Empowers Intelligence: Integrating Expertise of Software Engineering Agents

Generalization v.s. Memorization: Tracing Language Models’ Capabilities Back to Pretraining Data

SWE-Search: Enhancing Software Agents with Monte Carlo Tree Search and Iterative Refinement