Boxuan Li

Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces

arXiv 2026

ConsistCompose: Unified Multimodal Layout Control for Image Composition

arXiv 2025

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

arXiv 2024

TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks

arXiv 2024

Frank F. Xu

Graham Neubig

professor

Niklas Muennighoff

grad-student

Yufan Song

Ahson Saiyed

Akshay Anand

Alex Dimakis

Alexander G. Shaw

Amaad Martin

Andrew Lanpouthakoun