Explicit Evidence Grounding via Structured Inline Citation Generation
For AI systems requiring factual generation, this work identifies a critical bottleneck in evidence span identification that current LLMs fail to address.
FullCite generates structured inline citations linking claims to source documents and evidence spans. Evaluation on ASQA, BioASQ, and ExpertQA shows LLMs struggle with precise evidence span identification, highlighting a gap for faithful attributed QA.
As AI systems become more widely adopted, the demand for factual and faithful generation grows. Properly attributing information through citations becomes, therefore, crucial. This work introduces FullCite, a framework that, in contrast to most previous works, generates structured inline citations linking each claim to both its source document and supporting evidence. FullCite proposes three strategies to inline citation generation: prompt-based generation, constrained decoding over a citation grammar, and posthoc span alignment. Using three question answering benchmarks, namely, ASQA, BioASQ, and ExpertQA, we assess citation quality and faithfulness along three dimensions: document-level correctness, evidence span identification, and claim-citation faithfulness. Our evaluation shows that while LLMs are generally effective at identifying relevant documents, they struggle to identify the precise supporting spans within them. This gap suggests that achieving faithful attributed QA will require research to place greater emphasis on precise evidence span identification.