Lune

ACL Findings2025

Generate, Discriminate, Evolve: Enhancing Context Faithfulness via Fine-Grained Sentence-Level Self-Evolution

Kun Li, Tianhua Zhang, Yunxiang Li, Hongyin Luo, Abdalla Mohamed Salama Sayed Moustafa, Xixin Wu, James R. Glass, Helen M. Meng

Abstract

Improving context faithfulness in large language models is essential for developing trustworthy retrieval augmented generation systems and mitigating hallucinations, especially in long-form question answering (LFQA) tasks or scenarios involving knowledge conflicts. Existing methods either intervene LLMs only at inference without addressing their inherent limitations or overlook the potential for self-improvement. In this paper, we introduce GenDiE (Generate, Discriminate, Evolve), a novel self-evolving framework that enhances context faithfulness through finegrained sentence-level optimization. GenDiE combines both generative and discriminative training, equipping LLMs with self-generation and self-scoring capabilities to facilitate iterative self-evolution. This supports both data construction for model alignment and scoreguided search during inference. Furthermore, by treating each sentence in a response as an independent optimization unit, GenDiE effectively addresses the limitations of previous approaches that optimize at the holistic answer level, which may miss unfaithful details. Experiments on ASQA (in-domain LFQA) and ConFiQA (out-of-domain counterfactual QA) datasets demonstrate that GenDiE surpasses various baselines in both faithfulness and correctness, and exhibits robust performance for domain adaptation 1 .

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.

4 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • QLoRA: Efficient Finetuning of Quantized LLMs cites LoRA: Low-Rank Adaptation of Large Language Models
  • Teaching Large Language Models to Maintain Contextual Faithfulness via Synthetic Tasks and Reinforcement Learning cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  • Teaching Large Language Models to Maintain Contextual Faithfulness via Synthetic Tasks and Reinforcement Learning cites Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
  • RAG-Zeval: Enhancing RAG Responses Evaluator through End-to-End Reasoning and Ranking-Based Reinforcement Learning cites Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
  • Self-Rewarding Language Models cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  • REFO: Reinforced Evolutionary Faithfulness Optimization for Large Language Models cites LoRA: Low-Rank Adaptation of Large Language Models
  • REFO: Reinforced Evolutionary Faithfulness Optimization for Large Language Models cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  • REFO: Reinforced Evolutionary Faithfulness Optimization for Large Language Models cites Self-Rewarding Language Models
  • Copy-Paste to Mitigate Large Language Model Hallucinations cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  • Copy-Paste to Mitigate Large Language Model Hallucinations cites Teaching Large Language Models to Maintain Contextual Faithfulness via Synthetic Tasks and Reinforcement Learning
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers4

  1. Teaching Large Language Models to Maintain Contextual Faithfulness via Synthetic Tasks and Reinforcement LearningShuzheng Si, Haozhe Zhao, Cheng Gao, Yuzhuo Bai et al.AAAI 2026 · 4 citations
  2. Copy-Paste to Mitigate Large Language Model HallucinationsYongchao Long, Yingying Zhang, Xianbin Wen, Xian Wu et al.ICLR 2026 · 2 citations
  3. RAG-Zeval: Enhancing RAG Responses Evaluator through End-to-End Reasoning and Ranking-Based Reinforcement LearningKun Li, Yunxiang Li, Tianhua Zhang, Hongyin Luo et al.EMNLP 2025 · 1 citation
  4. REFO: Reinforced Evolutionary Faithfulness Optimization for Large Language ModelsYi Wang, Xiaqiang Tang, Keyu Hu, Haojie Lu et al.AAAI 2026

Builds on24

  1. LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
  2. Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning et al.NeurIPS 2023 · 10,924 citations
  3. QLoRA: Efficient Finetuning of Quantized LLMsTim Dettmers, Artidoro Pagnoni, Ari Holtzman, Luke ZettlemoyerNeurIPS 2023 · 5,863 citations
  4. Self-RAG: Learning to Retrieve, Generate, and Critique through Self-ReflectionAkari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil et al.ICLR 2024 · 1,798 citations
  5. Self-Rewarding Language ModelsWeizhe Yuan, Richard Yuanzhe Pang, Kyunghyun Cho, Xian Li et al.ICML 2024 · 569 citations

Related papers

  1. Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced OptimizationLei Huang, Xiaocheng Feng, Weitao Ma, Yuchun Fan et al.ACL 2025
  2. Improving Context Fidelity via Native Retrieval-Augmented ReasoningSuyuchen Wang, Jinlin Wang, Xinyu Wang, Shiqi Li et al.EMNLP 2025 · 1 citation
  3. Context-DPO: Aligning Language Models for Context-FaithfulnessBaolong Bi, Shaohan Huang, Yiwei Wang, Tianchi Yang et al.ACL Findings 2025 · 2 citations
  4. Exploiting Contextual Knowledge in LLMs through V-usable Information based Layer EnhancementXiaowei Yuan, Zhao Yang, Ziyang Huang, Yequan Wang et al.ACL 2025 · 3 citations
  5. FaithfulRAG: Fact-Level Conflict Modeling for Context-Faithful Retrieval-Augmented GenerationQinggang Zhang, Zhishang Xiang, Yilin Xiao, Le Wang et al.ACL 2025 · 18 citations
Dusk over the sea between two cliffs drawn in fine vertical lines