mlsanityData to 14 Oct 2026

CAGE: Cognitive Attribution Graphs for Faithful Inline Citation Generation in Long-Form Question Answering

arXiv 2607.24236 ↗First posted 27 Jul 2026cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

Long-form question answering increasingly relies on retrieved evidence to make LLM outputs verifiable, with inline citations tracing claims to source documents. However, existing systems often attach citations that are topically related but insufficient to support their claims. We identify attribution ambiguity as a structural challenge: end-to-end generation must implicitly resolve combinatorial claim--document assignments, obscuring evidential boundaries and increasing the risk of evidence-boundary overrun, where claims exceed cited support. To address this challenge, we propose CAGE (Cognitive Attribution Graphs for Citation Generation), a two-stage framework that introduces an explicit cognitive attribution map before answer generation. CAGE first trains a plug-and-play Cognitive Map Induction Model to construct answer-centered support subgraphs, aligning each semantic answer unit with supporting documents through explicit relations. A Structured Citation Reasoning Model then realizes these units as sentence-level claims with map-aligned citations. Experiments on ASQA, ELI5, and ExpertQA show that CAGE achieves state-of-the-art performance, demonstrating the effectiveness of attribution-space contraction and map-guided citation generation.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 25 of its 37 references that are in the map. First posted on arXiv 27 Jul 2026.

Builds on

25 of 37 references.

References · 25

  1. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  2. 2023Unifying Large Language Models and Knowledge Graphs: A Roadmapcited by 668
  3. 2023Enabling Large Language Models to Generate Text with Citationscited by 595
  4. 2019ELI5: Long Form Question Answeringcited by 575
  5. 2022ASQA: Factoid Questions Meet Long-Form Answerscited by 234
  6. 2025Large Language Model Agent: A Survey on Methodology, Applications and Challengescited by 205
  7. 2023ExpertQA: Expert-Curated Questions and Attributed Answerscited by 97
  8. 2024LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QAcited by 81
  9. 2023Large Language Models and Knowledge Graphs: Opportunities and Challengescited by 71
  10. 2024Attribute First, then Generate: Locally-attributable Grounded Text Generationcited by 54
  11. 2023Effective Large Language Model Adaptation for Improved Grounding and Citation Generationcited by 45
  12. 2024Measuring and Enhancing Trustworthiness of LLMs in RAG through Grounded Attributions and Learning to Refusecited by 34
13 more
  1. 2024Source-Aware Training Enables Knowledge Attribution in Language Modelscited by 30
  2. 2024Improving Attributed Text Generation of Large Language Models via Preference Learningcited by 23
  3. 2024Atomic Fact Decomposition Helps Attributed Question Answeringcited by 21
  4. 2023A Survey on Temporal Knowledge Graph Completion: Taxonomy, Progress, and Prospectscited by 14
  5. 2024Advancing Large Language Model Attribution through Self-Improvingcited by 14
  6. 2026Chaining the Evidence: Robust Reinforcement Learning for Deep Search Agents with Citation-Aware Rubric Rewardscited by 14
  7. 2026Relevant Is Not Warranted: Evidence-Force Calibration for Cited RAGcited by 14
  8. 2025LAQuer: Localized Attribution Queries in Content-grounded Generationcited by 13
  9. 2024Effective Instruction Parsing Plugin for Complex Logical Query Answering on Knowledge Graphscited by 8
  10. 2025Lessons from Training Grounded LLMs with Verifiable Rewardscited by 8
  11. 2024Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward Modelingcited by 6
  12. 2025Transparentize the Internal and External Knowledge Utilization in LLMs with Trustworthy Citationcited by 4
  13. 2025Prompting Large Language Models with Partial Knowledge for Answering Questions with Unseen Entitiescited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Trajectory-Guided Forget-Recover Network for Continual LLM Unlearningcited 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. 2025VeriCite: Towards Reliable Citations in Retrieval-Augmented Generation via Rigorous Verification
  3. 2026ProvenAI: Provenance-Native Traces of Evidence in Generated Answers
  4. 2026Why Retrieval-Augmented Generation Fails: A Graph Perspective
  5. 2025Concise and Sufficient Sub-Sentence Citations for Retrieval-Augmented Generation
  6. 2026Improving Attributed Long-form Question Answering with Intent Awareness
  7. 2026Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation
  8. 2026REFACT: Adaptive Fact Restatement for Compact and Faithful Chain-of-Thought Reasoning