LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA
read the original abstract
Though current long-context large language models (LLMs) have demonstrated impressive capacities in answering user questions based on extensive text, the lack of citations in their responses makes user verification difficult, leading to concerns about their trustworthiness due to their potential hallucinations. In this work, we aim to enable long-context LLMs to generate responses with fine-grained sentence-level citations, improving their faithfulness and verifiability. We first introduce LongBench-Cite, an automated benchmark for assessing current LLMs' performance in Long-Context Question Answering with Citations (LQAC), revealing considerable room for improvement. To this end, we propose CoF (Coarse to Fine), a novel pipeline that utilizes off-the-shelf LLMs to automatically generate long-context QA instances with precise sentence-level citations, and leverage this pipeline to construct LongCite-45k, a large-scale SFT dataset for LQAC. Finally, we train LongCite-8B and LongCite-9B using the LongCite-45k dataset, successfully enabling their generation of accurate responses and fine-grained sentence-level citations in a single output. The evaluation results on LongBench-Cite show that our trained models achieve state-of-the-art citation quality, surpassing advanced proprietary models including GPT-4o.
This paper has not been read by Pith yet.
Forward citations
Cited by 7 Pith papers
-
Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents?
REFLECT benchmark shows current LLM judges achieve below 55% accuracy detecting failures in evidence-based research agents, especially on evidence verification.
-
SPIRE: Structure-Preserving Interpretable Retrieval of Evidence
SPIRE presents a tree-structured retrieval method using subdocuments, paths, and dual contextualization that produces higher-quality and more diverse citations than passage-based baselines on HTML QA benchmarks.
-
What LLMs explain is not what they believe: Evaluating explanation sufficiency under models' own input beliefs
Proposes SCSuff metric for evaluating LLM explanation sufficiency via model-generated alternative inputs, showing explanations are typically insufficient and predictable from hidden states.
-
Do LLM Attribution Metrics Transfer? Auditing Retrieval-Augmented Generation Evaluation Across Datasets and Constructs
No automatic attribution scorer transfers across datasets in generated-answer attribution evaluation; per-dataset rankings invert and some drop to chance level, requiring target-dataset validation.
-
Citation Failure: Definition, Analysis and Efficient Mitigation
The paper defines citation failure separately from response failure, introduces the CITECONTROL benchmark to analyze relational effects on citation quality, and proposes the CITENTION framework that integrates generat...
-
Context Attribution with Multi-Armed Bandit Optimization
Formulates context attribution as a combinatorial multi-armed bandit problem solved via Linear Thompson Sampling to reduce LLM queries by up to 30% on QA benchmarks while matching existing attribution quality.
-
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
InternVL 2.5 is the first open-source MLLM to surpass 70% on the MMMU benchmark via model, data, and test-time scaling, with a 3.7-point gain from chain-of-thought reasoning.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.