PaperTrail: A Claim-Evidence Interface for Grounding Provenance in LLM-based Scholarly Q&A
This addresses the need for rigorous verification in scholarly domains by providing a more granular provenance mechanism, though it is incremental as it focuses on interface design rather than solving the underlying cognitive burden.
The paper tackled the problem of subtle errors in LLM-based scholarly QA systems by introducing PaperTrail, a claim-evidence interface for grounding provenance, and found that it significantly lowered participants' trust compared to a baseline but did not change their reliance on LLM-generated edits.
Large language models (LLMs) are increasingly used in scholarly question-answering (QA) systems to help researchers synthesize vast amounts of literature. However, these systems often produce subtle errors (e.g., unsupported claims, errors of omission), and current provenance mechanisms like source citations are not granular enough for the rigorous verification that scholarly domain requires. To address this, we introduce PaperTrail, a novel interface that decomposes both LLM answers and source documents into discrete claims and evidence, mapping them to reveal supported assertions, unsupported claims, and information omitted from the source texts. We evaluated PaperTrail in a within-subjects study with 26 researchers who performed two scholarly editing tasks using PaperTrail and a baseline interface. Our results show that PaperTrail significantly lowered participants' trust compared to the baseline. However, this increased caution did not translate to behavioral changes, as people continued to rely on LLM-generated scholarly edits to avoid a cognitively burdensome task. We discuss the value of claim-evidence matching for understanding LLM trustworthiness in scholarly settings, and present design implications for cognition-friendly communication of provenance information.