LFD: Layer Fused Decoding to Exploit External Knowledge in Retrieval-Augmented Generation

• URL: http://arxiv.org/abs/2508.19614v2
• Date: Thu, 23 Oct 2025 06:59:59 GMT
• Title: LFD: Layer Fused Decoding to Exploit External Knowledge in Retrieval-Augmented Generation
• Authors: Yang Sun, Zhiyong Xie, Dan Luo, Long Zhang, Liming Dong, Yunwei Zhao, Xixun Lin, Yanxiong Lu, Chenliang Li, Lixin Zou,
• Abstract summary: Retrieval-augmented generation (RAG) incorporates external knowledge into large language models (LLMs)<n>Recent empirical evidence demonstrates that injecting noise into retrieved relevant documents paradoxically facilitates exploitation of external knowledge and improves generation quality.<n>We propose Layer Fused Decoding (LFD), a simple decoding strategy that directly combines representations from an intermediate layer with final-layer decoding outputs.
• Score: 32.43047704369537
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Retrieval-augmented generation (RAG) incorporates external knowledge into large language models (LLMs), improving their adaptability to downstream tasks and enabling information updates. Surprisingly, recent empirical evidence demonstrates that injecting noise into retrieved relevant documents paradoxically facilitates exploitation of external knowledge and improves generation quality. Although counterintuitive and challenging to apply in practice, this phenomenon enables granular control and rigorous analysis of how LLMs integrate external knowledge. Therefore, in this paper, we intervene on noise injection and establish a layer-specific functional demarcation within the LLM: shallow layers specialize in local context modeling, intermediate layers focus on integrating long-range external factual knowledge, and deeper layers primarily rely on parametric internal knowledge. Building on this insight, we propose Layer Fused Decoding (LFD), a simple decoding strategy that directly combines representations from an intermediate layer with final-layer decoding outputs to fully exploit the external factual knowledge. To identify the optimal intermediate layer, we introduce an internal knowledge score (IKS) criterion that selects the layer with the lowest IKS value in the latter half of layers. Experimental results across multiple benchmarks demonstrate that LFD helps RAG systems more effectively surface retrieved context knowledge with minimal cost.

Related papers

• URaG: Unified Retrieval and Generation in Multimodal LLMs for Efficient Long Document Understanding [55.45331924836242]
  We present URaG, a framework that Unifies Retrieval and Generation within a single MLLM.<n>We show that URaG achieves state-of-the-art performance while reducing computational overhead by 44-56%.
  arXiv  Detail & Related papers  (2025-11-13T17:54:09Z)
• Beyond RAG vs. Long-Context: Learning Distraction-Aware Retrieval for Efficient Knowledge Grounding [5.353135097018941]
  Retrieval-Augmented Generation (RAG) is a framework for grounding Large Language Models (LLMs) in external, up-to-date information.<n>We propose LDAR (Learning Distraction-Aware Retrieval), an adaptive retriever that learns to retrieve contexts in a way that mitigates interference from distracting passages.
  arXiv  Detail & Related papers  (2025-09-26T04:40:42Z)
• DeepSieve: Information Sieving via LLM-as-a-Knowledge-Router [57.28685457991806]
  DeepSieve is an agentic RAG framework that incorporates information sieving via LLM-as-a-knowledge-router.<n>Our design emphasizes modularity, transparency, and adaptability, leveraging recent advances in agentic system design.
  arXiv  Detail & Related papers  (2025-07-29T17:55:23Z)
• Dynamic and Parametric Retrieval-Augmented Generation [17.311144793201652]
  Retrieval-Augmented Generation (RAG) has become a foundational paradigm for equipping large language models with external knowledge.<n>This tutorial delves into two rapidly growing and complementary research areas on RAG: Dynamic RAG and Parametric RAG.
  arXiv  Detail & Related papers  (2025-06-07T07:59:33Z)
• R1-Searcher++: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning [83.256752220849]
  Large Language Models (LLMs) are powerful but prone to hallucinations due to static knowledge.<n>We introduce R1-Searcher++, a framework designed to train LLMs to adaptively leverage both internal and external knowledge sources.<n>Our experiments demonstrate that R1-Searcher++ outperforms previous RAG and reasoning methods and achieves efficient retrieval.
  arXiv  Detail & Related papers  (2025-05-22T17:58:26Z)
• Unveiling Knowledge Utilization Mechanisms in LLM-based Retrieval-Augmented Generation [77.10390725623125]
  retrieval-augmented generation (RAG) is widely employed to expand their knowledge scope.<n>Since RAG has shown promise in knowledge-intensive tasks like open-domain question answering, its broader application to complex tasks and intelligent assistants has further advanced its utility.<n>We present a systematic investigation of the intrinsic mechanisms by which RAGs integrate internal (parametric) and external (retrieved) knowledge.
  arXiv  Detail & Related papers  (2025-05-17T13:13:13Z)
• Exploiting Contextual Knowledge in LLMs through V-usable Information based Layer Enhancement [20.183957585014042]
  We propose Context-aware Layer Enhancement (CaLE) to enhance the utilization of contextual knowledge in large language models.<n>CaLE strategically amplifies the growth of contextual information at an optimal layer, thereby enriching representations in the final layer.<n>Our experiments demonstrate that CaLE effectively improves context-faithful generation in Question-Answering tasks.
  arXiv  Detail & Related papers  (2025-04-22T06:42:22Z)
• ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation [91.20492150248106]
  We investigate the internal mechanisms behind unfaithful generation and identify a subset of mid-to-deep feed-forward networks (FFNs) that are disproportionately activated in such cases.<n>We propose Parametric Knowledge Muting through FFN Suppression (ParamMute), a framework that improves contextual faithfulness by suppressing the activation of unfaithfulness-associated FFNs.<n> Experimental results show that ParamMute significantly enhances faithfulness across both CoFaithfulQA and the established ConFiQA benchmark, achieving substantial reductions in reliance on parametric memory.
  arXiv  Detail & Related papers  (2025-02-21T15:50:41Z)
• FactorLLM: Factorizing Knowledge via Mixture of Experts for Large Language Models [50.331708897857574]
  We introduce FactorLLM, a novel approach that decomposes well-trained dense FFNs into sparse sub-networks without requiring any further modifications. FactorLLM achieves comparable performance to the source model securing up to 85% model performance while obtaining over a 30% increase in inference speed.
  arXiv  Detail & Related papers  (2024-08-15T16:45:16Z)
• Wiki-LLaVA: Hierarchical Retrieval-Augmented Generation for Multimodal LLMs [39.54891426369773]
  We focus on endowing such models with the capability of answering questions that require external knowledge. Our approach, termed Wiki-LLaVA, aims at integrating an external knowledge source of multimodal documents. We conduct extensive experiments on datasets tailored for visual question answering with external data and demonstrate the appropriateness of our approach.
  arXiv  Detail & Related papers  (2024-04-23T18:00:09Z)
• REAR: A Relevance-Aware Retrieval-Augmented Framework for Open-Domain Question Answering [115.72130322143275]
  REAR is a RElevance-Aware Retrieval-augmented approach for open-domain question answering (QA) We develop a novel architecture for LLM-based RAG systems, by incorporating a specially designed assessment module. Experiments on four open-domain QA tasks show that REAR significantly outperforms previous a number of competitive RAG approaches.
  arXiv  Detail & Related papers  (2024-02-27T13:22:51Z)
• BlendFilter: Advancing Retrieval-Augmented Large Language Models via Query Generation Blending and Knowledge Filtering [58.403898834018285]
  BlendFilter is a novel approach that elevates retrieval-augmented Large Language Models by integrating query generation blending with knowledge filtering. We conduct extensive experiments on three open-domain question answering benchmarks, and the findings clearly indicate that our innovative BlendFilter surpasses state-of-the-art baselines significantly.
  arXiv  Detail & Related papers  (2024-02-16T23:28:02Z)
• Accelerating LLaMA Inference by Enabling Intermediate Layer Decoding via Instruction Tuning with LITE [62.13435256279566]
  Large Language Models (LLMs) have achieved remarkable performance across a wide variety of natural language tasks. However, their large size makes their inference slow and computationally expensive. We show that it enables these layers to acquire 'good' generation ability without affecting the generation ability of the final layer.
  arXiv  Detail & Related papers  (2023-10-28T04:07:58Z)

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.