Pith. sign in

REVIEW 11 cited by

DeepRAG: Thinking to Retrieve Step by Step for Large Language Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.01142 v2 pith:DKQPLDWI submitted 2025-02-03 cs.AI cs.CLcs.IR

Xinyan Guan , Jiali Zeng , Fandong Meng , Chunlei Xin , Yaojie Lu , Hongyu Lin , Xianpei Han , Le Sun
show 1 more author
Jie Zhou
This is my paper · ORCID
classification cs.AIcs.CLcs.IR
keywords reasoningdeepragmodelsretrievalretrieval-augmentedstepaccuracyenhancing
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large Language Models (LLMs) have shown remarkable reasoning capabilities, while their practical applications are limited by severe factual hallucinations due to limitations in the timeliness, accuracy, and comprehensiveness of their parametric knowledge. Meanwhile, enhancing retrieval-augmented generation (RAG) with reasoning remains challenging due to ineffective task decomposition and redundant retrieval, which can introduce noise and degrade response quality. In this paper, we propose DeepRAG, a framework that models retrieval-augmented reasoning as a Markov Decision Process (MDP), enabling reasonable and adaptive retrieval. By iteratively decomposing queries, DeepRAG dynamically determines whether to retrieve external knowledge or rely on parametric reasoning at each step. Experiments show that DeepRAG improves retrieval efficiency and boosts answer accuracy by 26.4%, demonstrating its effectiveness in enhancing retrieval-augmented reasoning.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. PROGRESS: Coverage-guided RL to Train Search-augmented LLM Agent

    cs.AI 2026-08 conditional novelty 6.0 of 10

    PROGRESS adds a teacher-guided coverage reward to RL training, encouraging search-augmented LLMs to issue comprehensive, granular sub-queries and improving multi-hop QA accuracy by 2-5% absolute.

  2. Fishing Out Free Riders: Shapley-Based Reward Attribution for Parallel Reasoning via Reinforcement Learning

    cs.AI 2026-07 conditional novelty 6.0 of 10

    Rewarding each parallel reasoning path by Monte-Carlo-Shapley marginal contribution, scored by a generative reward model, lifts Pass@16 on AIME24/AIME25/AMC23 by 4-90% relative over Parallel-R1 with a fifth of the tra...

  3. MG$^2$-RAG: Multi-Granularity Graph for Multimodal Retrieval-Augmented Generation

    cs.IR 2026-04 unverdicted novelty 6.0 of 10

    MG²-RAG proposes a multi-granularity graph RAG framework that constructs hierarchical multimodal nodes via entity-driven visual grounding and performs structured retrieval, delivering SOTA results on four multimodal t...

  4. Optimizing RAG Rerankers with LLM Feedback via Reinforcement Learning

    cs.CL 2026-04 conditional novelty 6.0 of 10

    RRPO formulates document reranking as a sequential MDP and optimizes a pointwise reranker with PPO using LLM generation rewards and a reference-anchored deterministic baseline.

  5. HGMEM: Hypergraph-based Working Memory to Improve Multi-step RAG for Long-Context Complex Relational Modeling

    cs.CL 2025-12 conditional novelty 6.0 of 10

    A working memory represented as a hypergraph, whose hyperedges are updated, inserted, and progressively merged by the LLM, improves multi-step RAG on long-context sense-making benchmarks.

  6. Coordinating Search-Informed Reasoning and Reasoning-Guided Search in Claim Verification

    cs.AI 2025-06 conditional novelty 6.0 of 10

    A pair of reinforcement-learning-trained agents, one for reasoning and one for search, improves accuracy on multi-hop claim verification benchmarks.

  7. Inter-Passage Verification for Multi-evidence Multi-answer QA

    cs.CL 2025-05 conditional novelty 6.0 of 10

    RI2VER improves multi-answer question answering by independently reading many passages to collect candidates and then filtering them through generated verification questions backed by extra evidence.

  8. R-Search: Empowering LLM Reasoning with Search via Multi-Reward Reinforcement Learning

    cs.CL 2025-06 conditional novelty 5.0 of 10

    An RL framework for retrieval-augmented generation that rewards LLMs for well-timed searches, high-quality evidence, and final answers, improving multi-hop QA.

  9. Towards Better Instruction Following Retrieval Models

    cs.CL 2025-05 conditional novelty 5.0 of 10

    A new training corpus and embedding model improve instruction-following p-MRR by up to 9 points on FollowIR, MAIR, and Bright benchmarks.

  10. Chain-of-Thought Poisoning Attacks against R1-based Retrieval-Augmented Generation Systems

    cs.IR 2025-05 conditional novelty 5.0 of 10

    Wrapping erroneous knowledge in a reasoning model's own chain-of-thought template raises poisoning attack success on an R1-based RAG system by 10 percentage points over the strongest prior method in a 100-query MS MARCO test.

  11. Deep Research Agents: A Systematic Examination And Roadmap

    cs.AI 2025-06 conditional novelty 4.0 of 10

    A survey that organizes LLM-powered deep research agents into static versus dynamic workflows and single versus multi agent architectures, and reviews their benchmarks and open challenges.

Pith tools