MMSearch-R1 uses reinforcement learning to train multimodal models for on-demand multi-turn internet search with image and text tools, outperforming same-size RAG baselines and matching larger ones while cutting search calls by over 30%.
Open deep search: Democratizing search with open-source reasoning agents
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
roles
other 1polarities
unclear 1representative citing papers
The EDR system with outline reflection, dependency-controlled information flow, and evidence sufficiency criteria outperforms baselines on sales enablement and DeepResearch Bench by reducing premature stopping and improving report depth.
citing papers explorer
-
MMSearch-R1: Incentivizing LMMs to Search
MMSearch-R1 uses reinforcement learning to train multimodal models for on-demand multi-turn internet search with image and text tools, outperforming same-size RAG baselines and matching larger ones while cutting search calls by over 30%.
-
Don\'t Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination
The EDR system with outline reflection, dependency-controlled information flow, and evidence sufficiency criteria outperforms baselines on sales enablement and DeepResearch Bench by reducing premature stopping and improving report depth.
- The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes