Pith. sign in

REVIEW 11 cited by

DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of 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 2403.10081 v3 pith:VBRNBJUC submitted 2024-03-15 cs.CL cs.IR

Weihang Su , Yichen Tang , Qingyao Ai , Zhijing Wu , Yiqun Liu This is my paper · ORCID
classification cs.CLcs.IR
keywords generationretrievedraginretrievaldynamicinformationneedswhat
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Dynamic retrieval augmented generation (RAG) paradigm actively decides when and what to retrieve during the text generation process of Large Language Models (LLMs). There are two key elements of this paradigm: identifying the optimal moment to activate the retrieval module (deciding when to retrieve) and crafting the appropriate query once retrieval is triggered (determining what to retrieve). However, current dynamic RAG methods fall short in both aspects. Firstly, the strategies for deciding when to retrieve often rely on static rules. Moreover, the strategies for deciding what to retrieve typically limit themselves to the LLM's most recent sentence or the last few tokens, while the LLM's real-time information needs may span across the entire context. To overcome these limitations, we introduce a new framework, DRAGIN, i.e., Dynamic Retrieval Augmented Generation based on the real-time Information Needs of LLMs. Our framework is specifically designed to make decisions on when and what to retrieve based on the LLM's real-time information needs during the text generation process. We evaluate DRAGIN along with existing methods comprehensively over 4 knowledge-intensive generation datasets. Experimental results show that DRAGIN achieves superior performance on all tasks, demonstrating the effectiveness of our method. We have open-sourced all the code, data, and models in GitHub: https://github.com/oneal2000/DRAGIN/tree/main

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. MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio

    cs.CL 2026-06 unverdicted novelty 8.0 of 10

    MetaSyn is a stage-level benchmark of 442 meta-analyses showing LLM agents retrieve up to 90.9% of eligible studies but include at most 52.7% in their final reports.

  2. Retrieval Feedback Memory Enhancement Large Model Retrieval Generation Method

    cs.IR 2025-08 reject novelty 6.0 of 10

    RFM-RAG improves RAG-based QA by iteratively building an evidence pool and stopping via a learned sufficiency classifier.

  3. LLM-based Question-Answer Framework for Sensor-driven HVAC System Interaction

    cs.AI 2025-07 conditional novelty 6.0 of 10

    JARVIS, an LLM-based HVAC question-answering framework with an Expert-LLM, a parameterized SQL builder, and bottom-up planning, outperforms a text-to-SQL baseline and its own ablations on a small expert-curated dataset.

  4. EraRAG: Efficient and Incremental Retrieval Augmented Generation for Growing Corpora

    cs.IR 2025-06 conditional novelty 6.0 of 10

    EraRAG uses hyperplane-based locality-sensitive hashing to build a hierarchical retrieval graph whose affected regions only are re-summarized when new documents arrive, cutting update cost by up to an order of magnitude.

  5. Maximally-Informative Retrieval for State Space Model Generation

    cs.CL 2025-06 conditional novelty 6.0 of 10

    RICO ranks documents by how much they reduce an SSM's question perplexity, using gradient-document inner products, and matches BM25 while often beating E5 on answer quality without finetuning.

  6. GOSU: Retrieval-Augmented Generation with Global-Level Optimized Semantic Unit-Centric Framework

    cs.CL 2025-08 reject novelty 5.0 of 10

    GOSU globally merges semantic units from text chunks into a unit-centric knowledge graph and uses three-tier keyword retrieval to improve RAG generation quality, according to LLM-judge win rates.

  7. ARAG: Agentic Retrieval Augmented Generation for Personalized Recommendation

    cs.IR 2025-06 conditional novelty 5.0 of 10

    A four-agent LLM pipeline (user understanding, natural language inference, context summarization, and ranking) improves retrieval-augmented product recommendations on Amazon data by up to 42% in NDCG@5 over recency an...

  8. Reasoning with RAGged events: RAG-Enhanced Event Knowledge Base Construction and reasoning with proof-assistants

    cs.CL 2025-06 reject novelty 5.0 of 10

    On a single Thucydides corpus, strong LLMs extract more events without RAG, weak LLMs benefit from simple RAG but collapse under complex RAG, and a Coq translation is proposed to justify RAG-discovered event types.

  9. AugServe: Adaptive Request Scheduling for Augmented Large Language Model Inference Serving

    cs.CL 2025-12 conditional novelty 4.0 of 10

    An adaptive two-stage scheduler plus dynamic token batching improves SLO-satisfying throughput for tool-augmented LLM inference versus vLLM and InferCept in the reported experiments.

  10. CogEvo-Edu: Cognitive Evolution Educational Multi-Agent Collaborative System

    cs.AI 2025-11 conditional novelty 4.0 of 10

    A hierarchical multi-agent tutoring system that jointly evolves a student profile, a value-scored knowledge base, and a teaching policy reports large LLM-judged gains over static RAG and single-agent baselines.

  11. AdaDocVQA: Adaptive Framework for Long Document Visual Question Answering in Low-Resource Settings

    cs.CL 2025-08 conditional novelty 4.0 of 10

    A hybrid retrieval, data augmentation, and ensemble inference framework fine-tunes Qwen2.5-VL-72B to 59% on the Japanese LAVA document VQA benchmark.

Pith tools