lumos_web_agent_ground_iterative

🪄 Agent Lumos: Unified and Modular Training for Open-Source Language Agents

<p align="center"> 🌐<a href="https://allenai.github.io/lumos">[Website]</a> &nbsp; 📝<a href="https://arxiv.org/abs/2311.05657">[Paper]</a> &nbsp; 🤗<a href="https://huggingface.co/datasets?sort=trending&search=ai2lumos">[Data]</a> &nbsp; 🤗<a href="https://huggingface.co/models?sort=trending&search=ai2lumos">[Model]</a> &nbsp; 🤗<a href="https://huggingface.co/spaces/ai2lumos/lumos_data_demo">[Demo]</a> &nbsp; </p>

We introduce 🪄Lumos, Language Agents with Unified Formats, Modular Design, and Open-Source LLMs. Lumos unifies a suite of complex interactive tasks and achieves competitive performance with GPT-4/3.5-based and larger open-source agents.

Lumos has following features:

• 🧩 Modular Architecture:
  • 🧩 Lumos consists of planning, grounding, and execution modules built based on LLAMA-2-7B/13B and off-the-shelf APIs.
  • 🤗 Lumos utilizes a unified data format that encompasses multiple task types, thereby enabling the developed agent framework to conveniently support a range of interactive tasks.
• 🌍 Diverse Training Data:
  • 🌍 Lumos is trained with ~56K diverse high-quality subgoal/action annotations from ground-truth reasoning steps in existing benchmarks with GPT-4.
  • ⚒️ Lumos data can be instrumental for future research in developing open-source agents for complex interactive tasks.
• 🚀 Competitive Performance:
  • 🚀 Lumos is comparable or even beats GPT-series agents on web/complex QA tasks Mind2Web and HotpotQA, and larger open agents on math and multimodal tasks.
  • 🚀 Lumos exceeds contemporaneous agents that have been fine-tuned with in-domain HotpotQA, Mind2Web and ScienceQA annotations, such as FiReAct, AgentLM, and AutoAct.
  • 🚀 Lumos performs better than open agent baseline formulations including chain-of-thoughts and integrated training.
  • 🚀 Lumos surpasses larger open LLM agents and domain-specific agents on unseen tasks, WebShop and InterCode_SQL.

Model Overview

lumos_web_agent_ground_iterative is a grounding module checkpoint finetuned on web agent task in Lumos-Iterative (Lumos-I) formulation.

The training annotation is shown below:

Citation

If you find this work is relevant with your research, please feel free to cite our work!

@article{yin2023lumos,
  title={Agent Lumos: Unified and Modular Training for Open-Source Language Agents},
  author={Yin, Da and Brahman, Faeze and Ravichander, Abhilasha and Chandu, Khyathi and Chang, Kai-Wei and Choi, Yejin and Lin, Bill Yuchen},
  journal={arXiv preprint arXiv:2311.05657},
  year={2023}
}