Method Drift›Retrieval-augmented generation
DeepRAG
DeepRAG: Thinking to Retrieve Step by Step for Large Language ModelsRetrieval-augmented generation · 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superseded — cited as a baseline and beaten by newer methods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.
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats DeepRAG · CWQA [Qwen2.5-7B]
61.4 vs 60.9
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats DeepRAG · GPQA [Qwen2.5-7B]
64.6 vs 61.3
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats DeepRAG · FMT [Qwen2.5-7B]
68.3 vs 66.9
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats DeepRAG · CWQA [Llama 3.1-8B]
67.3 vs 62.3
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats DeepRAG · GPQA [Llama 3.1-8B]
71.3 vs 69.8
- MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
MCTS-RAG beats DeepRAG · FMT [Llama 3.1-8B]
73.8 vs 72.9
- Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
HGMem beats DeepRAG · Comprehensiveness [GPT-4o]
65.73 vs 63.62
- Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
HGMem beats DeepRAG · Diversity [GPT-4o]
69.74 vs 65.98
- Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
HGMem beats DeepRAG · Accuracy [GPT-4o]
55.00 vs 45.00
- Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
HGMem beats DeepRAG · Comprehensiveness [Qwen2.5-32B-Instruct]
64.18 vs 61.45
- Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
HGMem beats DeepRAG · Diversity [Qwen2.5-32B-Instruct]
66.51 vs 63.56
- Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
HGMem beats DeepRAG · Accuracy [Qwen2.5-32B-Instruct]
51.00 vs 44.00
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.