Update README.md
Browse files
README.md
CHANGED
@@ -14,3 +14,45 @@ pinned: false
|
|
14 |
### Paper:
|
15 |
|
16 |
### Github: https://github.com/allenai/lumos
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
### Paper:
|
15 |
|
16 |
### Github: https://github.com/allenai/lumos
|
17 |
+
|
18 |
+
# 🪄 Lumos: Learning Agents with Unified Data, Modular Design, and Open-Source LLMs
|
19 |
+
<p align="center">
|
20 |
+
<a href="https://allenai.github.io/lumos/">
|
21 |
+
<img src="https://img.shields.io/badge/🌐-Website-red">
|
22 |
+
</a>
|
23 |
+
<a href="">
|
24 |
+
<img src="https://img.shields.io/badge/📝-Paper-blue">
|
25 |
+
</a>
|
26 |
+
<a href="https://huggingface.co/datasets?sort=trending&search=ai2lumos">
|
27 |
+
<img src="https://img.shields.io/badge/🤗-Data-orange">
|
28 |
+
</a>
|
29 |
+
<a href="https://huggingface.co/models?sort=trending&search=ai2lumos">
|
30 |
+
<img src="https://img.shields.io/badge/🤗-Model-green">
|
31 |
+
</a>
|
32 |
+
</p>
|
33 |
+
|
34 |
+
🖋 **Authors:** [Da Yin](https://wadeyin9712.github.io/), [Faeze Brahman](https://fabrahman.github.io/), [Abhilasha Ravichander](https://www.cs.cmu.edu/~aravicha/), [Khyathi Chandu](https://www.cs.cmu.edu/~kchandu/), [Kai-Wei Chang](http://web.cs.ucla.edu/~kwchang/), [Yejin Choi](https://homes.cs.washington.edu/~yejin/), [Bill Yuchen Lin](https://yuchenlin.xyz/)
|
35 |
+
|
36 |
+
We introduce 🪄**Lumos**, Language Agents with **Unified** Data 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.
|
37 |
+
|
38 |
+
‼️ **Lumos** has following features:
|
39 |
+
* 🧩 **Modular Architecture**:
|
40 |
+
- **Lumos** consists of planning, grounding, and execution modules built based on LLAMA-2-7B.
|
41 |
+
* 🌍 **Diverse Training Data**:
|
42 |
+
- **Lumos** is trained with ~40K high-quality annotations from ground-truth reasoning steps in existing benchmarks with GPT-4.
|
43 |
+
* 🚀 **Competitive Performance**:
|
44 |
+
- 🚀 **Lumos** outperforms **GPT-4/3.5-based** agents on complex QA and web tasks, and **larger open agents** on maths tasks.
|
45 |
+
- 🚀 **Lumos** performs better than open agent baseline formulations including **chain-of-thoughts** and **integrated** training.
|
46 |
+
- 🚀 **Lumos** surpasses larger open LLM agents and domain-specific agents on an unseen task, WebShop.
|
47 |
+
|
48 |
+
## 🤩 Citation
|
49 |
+
|
50 |
+
If you find this work is relevant with your research, please feel free to cite our work!
|
51 |
+
```
|
52 |
+
@article{yin2023lumos,
|
53 |
+
title={Lumos: Language Agents with Unified Data Formats, Modular Design, and Open-Source LLMs},
|
54 |
+
author={Yin, Da and Brahman, Faeze and Ravichander, Abhilasha and Chandu, Khyathi and Chang, Kai-Wei and Choi, Yejin and Lin, Bill Yuchen},
|
55 |
+
year={2023}
|
56 |
+
}
|
57 |
+
```
|
58 |
+
|