mlsanityData to 7 Oct 2026

SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models

arXiv 2502.09604 ↗First posted 13 Feb 2025cs.CL cs.AI cs.LGcited by 19

On the map: LLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG

Abstract (arXiv)

We introduce SelfCite, a novel self-supervised approach that aligns LLMs to generate high-quality, fine-grained, sentence-level citations for the statements in their generated responses. Instead of only relying on costly and labor-intensive annotations, SelfCite leverages a reward signal provided by the LLM itself through context ablation: If a citation is necessary, removing the cited text from the context should prevent the same response; if sufficient, retaining the cited text alone should preserve the same response. This reward can guide the inference-time best-of-N sampling strategy to improve citation quality significantly, as well as be used in preference optimization to directly fine-tune the models for generating better citations. The effectiveness of SelfCite is demonstrated by increasing citation F1 up to 5.3 points on the LongBench-Cite benchmark across five long-form question answering tasks. The source code is available at https://github.com/facebookresearch/SelfCite

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Placed in Preference alignment for recommenders and RAG, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 36 of its 44 references that are in the map. First posted on arXiv 13 Feb 2025.

Builds on

36 of 44 references.

References · 36

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2017Proximal Policy Optimization Algorithmscited by 14,441
  3. 2024The Llama 3 Herd of Modelscited by 14,192
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  6. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  7. 2023Let's Verify Step by Stepcited by 3,598
  8. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  9. 2022Survey of Hallucination in Natural Language Generationcited by 2,056
  10. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,722
  11. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,378
  12. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,096
24 more
  1. 2022Scaling Laws for Reward Model Overoptimizationcited by 1,056
  2. 2024SimPO: Simple Preference Optimization with a Reference-Free Rewardcited by 951
  3. 2024Self-Rewarding Language Modelscited by 654
  4. 2023Enabling Large Language Models to Generate Text with Citationscited by 584
  5. 2021Efficient Attentions for Long Document Summarizationcited by 342
  6. 2023Trusting Your Evidence: Hallucinate Less with Context-aware Decodingcited by 319
  7. 2022Teaching language models to support answers with verified quotescited by 290
  8. 2022RARR: Researching and Revising What Language Models Say, Using Language Modelscited by 286
  9. 2016Rationalizing Neural Predictionscited by 281
  10. 2024Iterative Reasoning Preference Optimizationcited by 244
  11. 2024Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Mapscited by 140
  12. 2023Controlled Decoding from Language Modelscited by 133
  13. 2024Liger Kernel: Efficient Triton Kernels for LLM Trainingcited by 123
  14. 2017DuReader: a Chinese Machine Reading Comprehension Dataset from Real-world Applicationscited by 110
  15. 2024LongAlign: A Recipe for Long Context Alignment of Large Language Modelscited by 102
  16. 2024ContextCite: Attributing Model Generation to Contextcited by 93
  17. 2024LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QAcited by 78
  18. 2023Aligning Large Language Models through Synthetic Feedbackcited by 75
  19. 2024Attribute First, then Generate: Locally-attributable Grounded Text Generationcited by 54
  20. 2024Training Language Models to Generate Text with Citations via Fine-grained Rewardscited by 40
  21. 2024Peering into the Mind of Language Models: An Approach for Attribution in Contextual Question Answeringcited by 25
  22. 2024Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generationcited by 25
  23. 2024Advancing Large Language Model Attribution through Self-Improvingcited by 14
  24. 2023Unifying Corroborative and Contributive Attributions in Large Language Modelscited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 14; 2026: 5.

19 papers.

Sort:

By area:

  1. 2025A Survey of Context Engineering for Large Language Modelscited by 120
  2. 2025Search Arena: Analyzing Search-Augmented LLMscited by 30
  3. 2025LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contextscited by 25
  4. 2025Attributing Response to Context: A Jensen-Shannon Divergence Driven Mechanistic Study of Context Attribution in Retrieval-Augmented Generationcited by 12
  5. 2025MedPAIR: Measuring Physicians and AI Relevance Alignment in Medical Question Answeringcited by 5
  6. 2025Generation-Time vs. Post-hoc Citation: A Holistic Evaluation of LLM Attributioncited by 5
  7. 2026Influential Training Data Retrieval for Explaining Verbalized Confidence of LLMscited by 4
  8. 2025MaxShapley: Towards Incentive-compatible Generative Search with Fair Context Attributioncited by 3
  9. 2025TrustRAG: An Information Assistant with Retrieval Augmented Generationcited by 1
  10. 2025Cite Pretrain: Retrieval-Free Knowledge Attribution for Large Language Modelscited by 1
  11. 2025DePass: Unified Feature Attributing by Simple Decomposed Forward Passcited by 1
  12. 2025Short-Context Dominance: How Much Local Context Natural Language Actually Needs?cited by 1
7 more
  1. 2026Beyond Factual Accuracy: Evaluating Global Reasoning Integrity in RAG Systems with LogicScorecited by 1
  2. 2026ProvenAI: Provenance-Native Traces of Evidence in Generated Answerscited by 1
  3. 2025Evaluation Framework for Highlight Explanations of Context Utilisation in Language Modelscited by 0
  4. 2025MMLongCite: A Benchmark for Evaluating Fidelity of Long-Context Vision-Language Modelscited by 0
  5. 2025Citation Failure: Definition, Analysis and Efficient Mitigationcited by 0
  6. 2026TRUE: A Trustworthy Unified Explanation Framework for Large Language Model Reasoningcited by 0
  7. 2026Tell Me Why: Designing an Explainable LLM-based Dialogue System for Student Problem Behavior Diagnosiscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026How Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation
  2. 2025Concise and Sufficient Sub-Sentence Citations for Retrieval-Augmented Generation
  3. 2025VeriCite: Towards Reliable Citations in Retrieval-Augmented Generation via Rigorous Verification
  4. 2024Learning to Generate Answers with Citations via Factual Consistency Models
  5. 2024Citekit: A Modular Toolkit for Large Language Model Citation Generation
  6. 2026AtomCite: Verification and Correction of Supplied Page-Level Citations in Multi-Page Documents
  7. 2025SCALAR: Scientific Citation-based Live Assessment of Long-context Academic Reasoning
  8. 2025SSFO: Self-Supervised Faithfulness Optimization for Retrieval-Augmented Generation