LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG

• URL: http://arxiv.org/abs/2605.06285v1
• Date: Thu, 07 May 2026 13:56:13 GMT
• Title: LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG
• Abstract summary: LatentRAG is a novel framework that shifts both reasoning and retrieval from discrete language space to continuous latent space.<n>LatentRAG achieves performance comparable to explicit agentic RAG methods while reducing inference latency by approximately 90%.
• Score: 13.420568360763227
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Single-step retrieval-augmented generation (RAG) provides an efficient way to incorporate external information for simple question answering tasks but struggles with complex questions. Agentic RAG extends this paradigm by replacing single-step retrieval with a multi-step process, in which the large language model (LLM) acts as a search agent that generates intermediate thoughts and subqueries to iteratively interact with the retrieval system. This iterative process incurs substantial latency due to the autoregressive generation of lengthy thoughts and subqueries. To address this limitation, we propose LatentRAG, a novel framework that shifts both reasoning and retrieval from discrete language space to continuous latent space. Unlike existing explicit methods that generate natural language thoughts or subqueries token-by-token, LatentRAG produces latent tokens for thoughts and subqueries directly from the hidden states in a single forward pass. We align LLMs with dense retrieval models in the latent space, enabling retrieval over latent subquery tokens and supporting end-to-end joint optimization. To improve transparency and encourage semantically meaningful latent representations, we incorporate a parallel latent decoding mechanism that translates latent tokens back into natural language. Extensive experiments on seven benchmark datasets show that LatentRAG achieves performance comparable to explicit agentic RAG methods while reducing inference latency by approximately 90%, substantially narrowing the latency gap with traditional single-step RAG.

Related papers

• Generative Universal Multimodal Retrieval with Dual-role Identifiers [1.0981910170846492]
  We propose DrIG, a novel Generative framework for universal multimodal retrieval featuring Dual-role Identifiers.<n>DrIG consistently outperforms state-of-the-art generative multimodal baselines across diverse tasks, while hybrid reranking achieves a favorable efficiency-effectiveness trade-off against strong dense retrievers.
  arXiv  Detail & Related papers  (2026-08-13T09:10:09Z)
• Latent Abstraction for Retrieval-Augmented Generation [4.6096940605642915]
  Existing systems rely on generating natural language queries at each hop.<n>We propose textbfLAnR (Latent Abstraction for RAG), a unified framework in which a single LLM jointly performs encoding, retrieval, and generation entirely within its own latent space.
  arXiv  Detail & Related papers  (2026-04-20T06:26:13Z)
• Transforming External Knowledge into Triplets for Enhanced Retrieval in RAG of LLMs [55.78708003681562]
  Retrieval-Augmented Generation (RAG) mitigates hallucination in large language models (LLMs) by incorporating external knowledge during generation.<n>Existing RAG approaches typically retrieve and unstructured text fragments as context.<n>We propose Tri-RAG, a structured triplet-based retrieval framework that improves retrieval efficiency through reasoning-aligned context construction.
  arXiv  Detail & Related papers  (2026-04-14T11:36:29Z)
• PLUME: Latent Reasoning Based Universal Multimodal Embedding [52.35354073629127]
  Universal multimodal embedding (UME) maps heterogeneous inputs into a shared retrieval space with a single model.<n>Recent approaches improve UME by generating explicit chain-of-thought (CoT) rationales before extracting embeddings.<n>We propose PLUME, a latent reasoning framework that advances UME by replacing verbalized CoT with a short autoregressive rollout of continuous latent states.
  arXiv  Detail & Related papers  (2026-04-02T14:04:53Z)
• LaSER: Internalizing Explicit Reasoning into Latent Space for Dense Retrieval [74.72139580745511]
  LaSER is a novel self-distillation framework that internalizes explicit reasoning into the latent space of retrievers.<n>Our method successfully combines the reasoning depth of explicit CoT pipelines with the inference efficiency of standard dense retrievers.
  arXiv  Detail & Related papers  (2026-03-02T04:11:18Z)
• Agentic Spatio-Temporal Grounding via Collaborative Reasoning [80.83158605034465]
  Temporal Video Grounding aims to retrieve thetemporal tube of a target object or person in a video given a text query.<n>We propose the Agentic Spatio-Temporal Grounder (ASTG) framework for the task of STVG towards an open-world and training-free scenario.<n>Specifically, two specialized agents SRA (Spatial Reasoning Agent) and TRA (Temporal Reasoning Agent) constructed leveraging on modern Multimoal Large Language Models (MLLMs)<n>Experiments on popular benchmarks demonstrate the superiority of the proposed approach where it outperforms existing weakly-supervised and zero-shot approaches by a margin
  arXiv  Detail & Related papers  (2026-02-10T10:16:27Z)
• M-RAG: Making RAG Faster, Stronger, and More Efficient [15.147969102210759]
  We propose M-RAG, a novel Chunk-free retrieval strategy for large language models.<n>M-RAG extracts structured, k-v decomposition meta-markers, with a lightweight, intent-aligned retrieval key.<n>We show that M-RAG retrieves more answer-friendly evidence with high efficiency.
  arXiv  Detail & Related papers  (2026-01-06T15:14:54Z)
• REFRAG: Rethinking RAG based Decoding [67.4862300145604]
  REFRAG is an efficient decoding framework that compresses, senses, and expands to improve latency in RAG applications.<n>We provide rigorous validation of REFRAG across diverse long-context tasks, including RAG, multi-turn conversations, and long document summarization.
  arXiv  Detail & Related papers  (2025-09-01T03:31:44Z)
• Accelerating Adaptive Retrieval Augmented Generation via Instruction-Driven Representation Reduction of Retrieval Overlaps [16.84310001807895]
  This paper introduces a model-agnostic approach that can be applied to A-RAG methods.<n>Specifically, we use cache access and parallel generation to speed up the prefilling and decoding stages respectively.
  arXiv  Detail & Related papers  (2025-05-19T05:39:38Z)
• Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection [74.51523859064802]
  We introduce a new framework called Self-Reflective Retrieval-Augmented Generation (Self-RAG) Self-RAG enhances an LM's quality and factuality through retrieval and self-reflection. It significantly outperforms state-of-the-art LLMs and retrieval-augmented models on a diverse set of tasks.
  arXiv  Detail & Related papers  (2023-10-17T18:18:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.