R^2ec: Towards Large Recommender Models with Reasoning
Paper • 2505.16994 • Published • 2
How to use from
SGLangUse Docker images
docker run --gpus all \
--shm-size 32g \
-p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HF_TOKEN=<secret>" \
--ipc=host \
lmsysorg/sglang:latest \
python3 -m sglang.launch_server \
--model-path "dd101bb/gemma-2b-cds" \
--host 0.0.0.0 \
--port 30000# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "dd101bb/gemma-2b-cds",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'Quick Links
R2ec: Towards Large Recommender Models with Reasoning
R2ec is a large recommender model with reasoning, generating both natural language rationales and ranked item predictions. The model is fine-tuned with reinforcement learning to enhance its reasoning capabilities for more effective recommendations.
Sources
- Repository: https://github.com/YRYangang/RRec
- Paper: https://arxiv.org/abs/2505.16994
Licence
This code repository is licensed under MIT License. The use of R2ec models is also subject to MIT License. R2ec series support commercial use and distillation.
Citation
@misc{you2025r2ec,
title={$\text{R}^2\text{ec}$: Towards Large Recommender Models with Reasoning},
author={Runyang You and Yongqi Li and Xinyu Lin and Xin Zhang and Wenjie Wang and Wenjie Li and Liqiang Nie},
year={2025},
eprint={2505.16994},
archivePrefix={arXiv},
primaryClass={cs.IR}
}
Install from pip and serve model
# Install SGLang from pip: pip install sglang# Start the SGLang server: python3 -m sglang.launch_server \ --model-path "dd101bb/gemma-2b-cds" \ --host 0.0.0.0 \ --port 30000# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "dd101bb/gemma-2b-cds", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'