Skip to content
@xlang-ai

XLANG Lab

Developing embodied AI agents that empower users to use language to interact with digital and physical environments to carry out real-world tasks.

Welcome to the Executable Language Grounding (XLANG) Lab! We are part of the HKU NLP Group at the University of Hong Kong. XLang focuses on building language model agents that transform (“grounding”) language instructions into code or actions executable in real-world environments, including databases (data agent), web applications (plugins/web agent), and the physical world (robotic agent) etc,. It lies at the heart of language model agents or natural language interfaces that can interact with and learn from these real-world environments to facilitate human interaction with data analysis, web applications, and robotic instruction through conversation. Recent advances in XLang incorporate techniques such as LLM + external tools, code generation, semantic parsing, and dialog or interactive systems.

Pinned Loading

  1. OSWorld OSWorld Public

    [NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

    Python 2.3k 331

  2. aguvis aguvis Public

    [ICML2025] Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction

    Python 370 26

  3. OpenAgents OpenAgents Public

    [COLM 2024] OpenAgents: An Open Platform for Language Agents in the Wild

    Python 4.6k 502

  4. instructor-embedding instructor-embedding Public

    [ACL 2023] One Embedder, Any Task: Instruction-Finetuned Text Embeddings

    Python 2k 156

  5. text2reward text2reward Public

    [ICLR 2024 Spotlight] Text2Reward: Reward Shaping with Language Models for Reinforcement Learning

    Jupyter Notebook 187 12

  6. DS-1000 DS-1000 Public

    [ICML 2023] Data and code release for the paper "DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation".

    Python 258 27

Repositories

Showing 10 of 26 repositories
  • OSWorld Public

    [NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

    xlang-ai/OSWorld’s past year of commit activity
    Python 2,321 Apache-2.0 331 120 1 Updated Nov 19, 2025
  • OpenCUA Public

    OpenCUA: Open Foundations for Computer-Use Agents

    xlang-ai/OpenCUA’s past year of commit activity
    Python 573 MIT 65 15 0 Updated Nov 11, 2025
  • Spider2 Public

    [ICLR 2025 Oral] Spider 2.0: Evaluating Language Models on Real-World Enterprise Text-to-SQL Workflows

    xlang-ai/Spider2’s past year of commit activity
    HTML 652 MIT 106 70 3 Updated Nov 7, 2025
  • OSWorld-G Public

    [NeurIPS 2025 Spotlight] Scaling Computer-Use Grounding via UI Decomposition and Synthesis

    xlang-ai/OSWorld-G’s past year of commit activity
    TypeScript 130 6 9 1 Updated Nov 6, 2025
  • VideoAgentTrek Public

    The official repo of VideoAgentTrek

    xlang-ai/VideoAgentTrek’s past year of commit activity
    Python 32 MIT 3 1 0 Updated Oct 24, 2025
  • xlang-ai.github.io Public

    The official website of xlang.ai

    xlang-ai/xlang-ai.github.io’s past year of commit activity
    TypeScript 4 0 0 0 Updated Sep 27, 2025
  • BRIGHT Public

    [ICLR 2025] BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrieval

    xlang-ai/BRIGHT’s past year of commit activity
    Python 178 CC-BY-4.0 20 7 0 Updated Sep 13, 2025
  • AgentNetTool Public

    This is the official code base of AgentNetTool in OpenCUA. Website: https://opencua.xlang.ai/

    xlang-ai/AgentNetTool’s past year of commit activity
    TypeScript 28 MIT 7 1 0 Updated Sep 3, 2025
  • computer-agent-arena Public

    Computer Agent Arena: Test & compare AI agents in real desktop apps & web environments. Code/data coming soon!

    xlang-ai/computer-agent-arena’s past year of commit activity
    50 Apache-2.0 4 1 0 Updated Apr 7, 2025
  • aguvis Public

    [ICML2025] Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction

    xlang-ai/aguvis’s past year of commit activity
    Python 370 26 24 0 Updated Mar 7, 2025