Binyuan Hui

BIRD-INTERACT: Re-imagining Text-to-SQL Evaluation via Lens of Dynamic Interactions

VideoAgentTrek: Computer-Use Pretraining from Unlabeled Videos

PlotCraft: Pushing the Limits of LLMs for Complex and Interactive Data Visualization

MegaFlow: Large-Scale Distributed Orchestration System for the Agentic Era

SWE-RM: Execution-free Feedback for Software Engineering Agents

WorldPM: Understanding Scaling Patterns in Human Preference Modeling

C$^3$-Bench: Evaluating and Achieving Controllable Code Completion in Code LLM

BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution

Reinforcement Learning for Symbolic Graphics Code with Visual Feedback

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

OLMoE: Open Mixture-of-Experts Language Models

SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications

Parallel Scaling Law for Language Models

DEEM: Diffusion models serve as the eyes of large language models for image perception

OpenHands: An Open Platform for AI Software Developers as Generalist Agents

Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluation

Synthesizing Software Engineering Data in a Test-Driven Manner