qingfei1 commited on
Commit
13b052a
·
verified ·
1 Parent(s): 1a13032

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +25 -0
README.md ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ datasets:
5
+ - qingfei1/R-Search_datasets
6
+ license: apache-2.0
7
+ ---
8
+ # R-Search: Empowering LLM Reasoning with Search via Multi-Reward Reinforcement Learning
9
+
10
+ <p align="center">
11
+ 🤗 <a href="https://huggingface.co/datasets/qingfei1/R-Search_datasets" target="_blank">[R-Search Datasets] </a> • 💻 <a href="https://github.com/QingFei1/R-Search" target="_blank">[Github Repo]</a>
12
+ </p>
13
+
14
+ **R-Search** is a novel reinforcement learning framework for reasoning–search integration. It enables LLMs to autonomously perform multi-step reasoning with deep search interaction, and to learn optimal reasoning–search trajectories via multi-reward signals, substantially improving performance on complex logic- and knowledge-intensive tasks.
15
+
16
+ ## Trained Models
17
+
18
+ We open-sourced the following models trained only on the 2wikimultihopqa training set:
19
+
20
+ |Model|Huggingface Repo|Description|
21
+ |---|---|---|
22
+ |**R-Search-7b-grpo**| [🤗 Huggingface Repo](https://huggingface.co/qingfei1/R-Search-7b-grpo) | Trained **Qwen2.5-7B-Instruct** using the GRPO algorithm |
23
+ |**R-Search-3b-grpo**| [🤗 Huggingface Repo](https://huggingface.co/qingfei1/R-Search-3b-grpo) | Trained **Qwen2.5-3B-Instruct** using the GRPO algorithm |
24
+ |**R-Search-7b-ppo**| [🤗 Huggingface Repo](https://huggingface.co/qingfei1/R-Search-7b-ppo) | Trained **Qwen2.5-7B-Instruct** using the PPO algorithm |
25
+ |**R-Search-3b-ppo**| [🤗 Huggingface Repo](https://huggingface.co/qingfei1/R-Search-3b-ppo) | Trained **Qwen2.5-3B-Instruct** using the PPO algorithm |