EMNLP Findings2025
Uplift-RAG: Uplift-Driven Knowledge Preference Alignment for Retrieval-Augmented Generation
Changle Qu, Sunhao Dai, Hengyi Cai, Yiyang Cheng, Jun Xu, Shuaiqiang Wang, Dawei Yin
Abstract
Retrieval-augmented generation (RAG) has proven effective in enhancing the knowledge coverage of large language models (LLMs) and mitigating hallucinations by incorporating external retrieved documents. However, documents deemed relevant by the retriever are not necessarily helpful for answer generation, and including misleading information can even degrade performance. Existing efforts to estimate document utility often rely on downstream generation performance, which conflates the influence of external documents with the intrinsic knowledge of the LLM, thereby obscuring the actual contribution of the retrieved content. To address this, this paper proposes Uplift-RAG, a uplift-driven knowledge preference alignment framework for RAG. Specifically, we first propose an uplift-based definition of document utility that quantifies each document's marginal benefit over the LLM's internal knowledge. We then optimize the reranker with three alignment objectives to identify and prioritize documents based on their uplift. This enables dynamic selection of documents that address the LLM's knowledge gaps, going beyond fixed top-k selection, while reducing reference redundancy and the computational overhead of the LLM's input. Extensive experiments demonstrate the effectiveness of Uplift-RAG.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Self-Knowledge Guided Retrieval Augmentation for Large Language Models cites Dense Passage Retrieval for Open-Domain Question Answering
- Self-Knowledge Guided Retrieval Augmentation for Large Language Models cites Retrieval Augmented Language Model Pre-Training
- Query Rewriting in Retrieval-Augmented Large Language Models cites Dense Passage Retrieval for Open-Domain Question Answering
- Query Rewriting in Retrieval-Augmented Large Language Models cites Retrieval Augmented Language Model Pre-Training
- Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation cites Dense Passage Retrieval for Open-Domain Question Answering
- Learning to Retrieve from Agent Trajectories cites Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation
Cited by top-tier papers1
Builds on14
- Retrieval Augmented Language Model Pre-TrainingKelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat et al.ICML 2020 · 2,937 citations
- Query Rewriting in Retrieval-Augmented Large Language ModelsXinbei Ma, Yeyun Gong, Pengcheng He, Hai Zhao et al.EMNLP 2023 · 191 citations
- Dense Passage Retrieval for Open-Domain Question AnsweringVladimir Karpukhin, Barlas Oguz, Sewon Min, Patrick Lewis et al.EMNLP 2020 · 142 citations
- Self-Knowledge Guided Retrieval Augmentation for Large Language ModelsYile Wang, Peng Li, Maosong Sun, Yang LiuEMNLP Findings 2023 · 62 citations
- Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented GenerationGuanting Dong, Yutao Zhu, Chenghao Zhang, Zechen Wang et al.WWW 2025 · 44 citations
Related papers
- Incentivizing Retrieval-Augmented Generation via Inner Adaptive Context SelectionChenxu Cui, Lin Shen, Haihui Fan, Sa Zhu et al.SIGIR 2026
- GainRAG: Preference Alignment in Retrieval-Augmented Generation through Gain Signal SynthesisYi Jiang, Sendong Zhao, Jianbo Li, Haochun Wang et al.ACL 2025
- Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented GenerationPengyue Jia, Derong Xu, Xiaopeng Li, Zhaocheng Du et al.ACL Findings 2025 · 5 citations
- InfoGain-RAG: Boosting Retrieval-Augmented Generation through Document Information Gain-based Reranking and FilteringZihan Wang, Zihan Liang, Zhou Shao, Yufei Ma et al.EMNLP 2025 · 4 citations
- DynamicRAG: Leveraging Outputs of Large Language Model as Feedback for Dynamic Reranking in Retrieval-Augmented GenerationJiashuo Sun, Xianrui Zhong, Sizhe Zhou, Jiawei HanNeurIPS 2025 · 19 citations
