RUC-NLPIR/Search-o1

🔍 Search-o1: Agentic Search-Enhanced Large Reasoning Models [EMNLP 2025]

50
/ 100
Established

Integrates agentic retrieval-augmented generation (RAG) with a "Reason-in-Documents" module that allows reasoning models to dynamically trigger web searches during inference when encountering knowledge gaps, then seamlessly incorporate retrieved documents back into the reasoning chain. Supports multiple reasoning benchmarks (GPQA, MATH, AMC, AIME, code tasks) and open-domain QA through modular pipelines that enable batch generation with interleaved search across different LRM backbones.

1,182 stars.

No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

1,182

Forks

104

Language

Python

License

MIT

Last pushed

Nov 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/RUC-NLPIR/Search-o1"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.