# Generation-Time vs. Post-hoc Citation: A Holistic Evaluation of LLM Attribution

arXiv: https://arxiv.org/abs/2509.21557 · HTML: https://mlsanity.com/paper/2509.21557/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 34 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 25 of 25)

- [FEVER: a large-scale dataset for Fact Extraction and VERification](https://mlsanity.com/paper/1803.05355/) (2018)
- [Enabling Large Language Models to Generate Text with Citations](https://mlsanity.com/paper/2305.14627/) (2023)
- [ELI5: Long Form Question Answering](https://mlsanity.com/paper/1907.09190/) (2019)
- [Fact or Fiction: Verifying Scientific Claims](https://mlsanity.com/paper/2004.14974/) (2020)
- [ASQA: Factoid Questions Meet Long-Form Answers](https://mlsanity.com/paper/2204.06092/) (2022)
- [Trustworthy AI: From Principles to Practices](https://mlsanity.com/paper/2110.01167/) (2021)
- [A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications](https://mlsanity.com/paper/1804.09635/) (2018)
- [LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA](https://mlsanity.com/paper/2409.02897/) (2024)
- [QAMPARI: An Open-domain Question Answering Benchmark for Questions with Many Answers from Multiple Paragraphs](https://mlsanity.com/paper/2205.12665/) (2022)
- [Training Language Models to Generate Text with Citations via Fine-grained Rewards](https://mlsanity.com/paper/2402.04315/) (2024)
- [A Reasoning-Focused Legal Retrieval Benchmark](https://mlsanity.com/paper/2505.03970/) (2025)
- [Learning to Plan and Generate Text with Citations](https://mlsanity.com/paper/2404.03381/) (2024)
- [Citation-Enhanced Generation for LLM-based Chatbots](https://mlsanity.com/paper/2402.16063/) (2024)
- [Peering into the Mind of Language Models: An Approach for Attribution in Contextual Question Answering](https://mlsanity.com/paper/2405.17980/) (2024)
- [CiteBench: A benchmark for Scientific Citation Text Generation](https://mlsanity.com/paper/2212.09577/) (2022)
- [Ground Every Sentence: Improving Retrieval-Augmented LLMs with Interleaved Reference-Claim Generation](https://mlsanity.com/paper/2407.01796/) (2024)
- [SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models](https://mlsanity.com/paper/2502.09604/) (2025)
- [Local Citation Recommendation with Hierarchical-Attention Text Encoder and SciBERT-based Reranking](https://mlsanity.com/paper/2112.01206/) (2021)
- [Learning to Generate Answers with Citations via Factual Consistency Models](https://mlsanity.com/paper/2406.13124/) (2024)
- [Establishing and Evaluating Trustworthy AI: Overview and Research Challenges](https://mlsanity.com/paper/2411.09973/) (2024)
- [Abstention vs. Hallucination: Benchmarking LLM Source Attribution for Scientific Citations](https://mlsanity.com/paper/2405.02228/) (2024)
- [CiteFix: Enhancing RAG Accuracy Through Post-Processing Citation Correction](https://mlsanity.com/paper/2504.15629/) (2025)
- [CiteBART: Learning to Generate Citations for Local Citation Recommendation](https://mlsanity.com/paper/2412.17534/) (2024)
- [A Conceptual Framework for AI-based Decision Systems in Critical Infrastructures](https://mlsanity.com/paper/2504.16133/) (2025)
- [Recommending Missed Citations Identified by Reviewers: A New Task, Dataset and Baselines](https://mlsanity.com/paper/2403.01873/) (2024)

## Built on by (most cited first, 6 of 6)

- [Cited but Not Verified: Parsing and Evaluating Source Attribution in LLM Deep Research Agents](https://mlsanity.com/paper/2605.06635/) (2026)
- [Detecting and Correcting Reference Hallucinations in Commercial LLMs and Deep Research Agents](https://mlsanity.com/paper/2604.03173/) (2026)
- [Citation-Closure Retrieval and Per-Rule Attribution for Real-World Regulatory Compliance Question Answering](https://mlsanity.com/paper/2605.29742/) (2026)
- [MMLongCite: A Benchmark for Evaluating Fidelity of Long-Context Vision-Language Models](https://mlsanity.com/paper/2510.13276/) (2025)
- [Verified Misguidance: Measuring Structural Citation Failures in Search-Augmented LLMs](https://mlsanity.com/paper/2605.28565/) (2026)
- [AtomCite: Verification and Correction of Supplied Page-Level Citations in Multi-Page Documents](https://mlsanity.com/paper/2609.05802/) (2026)

## Abstract (arXiv)

Trustworthy Large Language Models (LLMs) must cite human-verifiable sources in high-stakes domains such as healthcare, law, academia, and finance, where even small errors can have severe consequences. Practitioners and researchers face a choice: let models generate citations during decoding, or let models draft answers first and then attach appropriate citations. To clarify this choice, we introduce two paradigms: Generation-Time Citation (G-Cite), which produces the answer and citations in one pass, and Post-hoc Citation (P-Cite), which adds or verifies citations after drafting. We conduct a comprehensive evaluation from zero-shot to advanced retrieval-augmented methods across four popular attribution datasets and provide evidence-based recommendations that weigh trade-offs across use cases. Our results show a consistent trade-off between coverage and citation correctness, with retrieval as the main driver of attribution quality in both paradigms. P-Cite methods achieve high coverage with competitive correctness and moderate latency, whereas G-Cite methods prioritize precision at the cost of coverage and speed. We recommend a retrieval-centric, P-Cite-first approach for high-stakes applications, reserving G-Cite for precision-critical settings such as strict claim verification. Our codes and human evaluation results are available at https://anonymous.4open.science/r/Citation_Paradigms-BBB5/