Update README.md
Browse files
README.md
CHANGED
|
@@ -29,4 +29,28 @@ Here's the user query:
|
|
| 29 |
<|im_start|>assistant
|
| 30 |
Let me write the SQL query with reasoning.
|
| 31 |
<think>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
```
|
|
|
|
| 29 |
<|im_start|>assistant
|
| 30 |
Let me write the SQL query with reasoning.
|
| 31 |
<think>
|
| 32 |
+
```
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
# DeepRetrieval
|
| 36 |
+
## Overview
|
| 37 |
+
|
| 38 |
+
DeepRetrieval is a novel approach that uses reinforcement learning (RL) to train Large Language Models (LLMs) for query generation without requiring supervised data. Instead of relying on expensive human-annotated or distilled reference queries, DeepRetrieval enables LLMs to learn through direct trial and error, using retrieval metrics as rewards.
|
| 39 |
+
## Key Features
|
| 40 |
+
|
| 41 |
+
- **No Supervision Required**: Eliminates the need for expensive human-annotated or distilled reference queries
|
| 42 |
+
- **RL-Based Framework**: Uses reinforcement learning to optimize query generation directly for retrieval performance
|
| 43 |
+
- **State-of-the-Art Performance**: Achieves remarkable results across diverse retrieval tasks
|
| 44 |
+
|
| 45 |
+
Please view our [GitHub page](https://github.com/pat-jj/DeepRetrieval) for instructions.
|
| 46 |
+
|
| 47 |
+
[DeepRetrieval Paper](arxiv.org/abs/2503.00223)
|
| 48 |
+
```
|
| 49 |
+
@article{jiang2025deepretrievalhackingrealsearch,
|
| 50 |
+
title={DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learning},
|
| 51 |
+
author={Pengcheng Jiang and Jiacheng Lin and Lang Cao and Runchu Tian and SeongKu Kang and Zifeng Wang and Jimeng Sun and Jiawei Han},
|
| 52 |
+
year={2025},
|
| 53 |
+
journal = {arXiv preprint arXiv: 2503.00223},
|
| 54 |
+
url={https://arxiv.org/abs/2503.00223}
|
| 55 |
+
}
|
| 56 |
```
|