Retrieval-augmented generation
DeepRAG
DeepRAG: Thinking to Retrieve Step by Step for Large Language Models
Superseded baseline#91 of 1,179 most-superseded · first seen Feb 3, 2025
Superseded — cited as a baseline and beaten by newer methods
0 papers critique it · 2 beat it on benchmarks
Beaten on benchmarks
Head-to-head results where a newer method reports beating DeepRAG. Values are copied from the source paper's tables — verify against the cited paper.
HGMem beats DeepRAG
55.00 vs 45.00
MCTS-RAG beats DeepRAG
67.3 vs 62.3
CWQA · [Llama 3.1-8B]
MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
What to use instead
Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.