From Binary Groundedness to Support Relations: Towards a Reader-Centred Taxonomy for Comprehension of AI Output

TL;DR

This paper proposes a reader-centred taxonomy of support relations for AI-generated answers, aiming to improve understanding and evaluation of how models reformulate source evidence.

Contribution

It introduces a novel taxonomy framework based on linguistic and philosophical insights to better characterize grounding in AI outputs.

Findings

Framework enables more nuanced provenance communication.

Proposed benchmark and annotation protocol for evaluation.

Enhances interpretability of AI-generated support relations.

Abstract

Generative AI tools often answer questions using source documents, e.g., through retrieval augmented generation. Current groundedness and hallucination evaluations largely frame the relationship between an answer and its sources as binary (the answer is either supported or unsupported). However, this obscures both the syntactic moves (e.g., direct quotation vs. paraphrase) and the interpretive moves (e.g., induction vs. deduction) performed when models reformulate evidence into an answer. This limits both benchmarking and user-facing provenance interfaces. We propose the development of a reader-centred taxonomy of grounding as a set of support relations between generated statements and source documents. We explain how this might be synthesised from prior research in linguistics and philosophy of language, and evaluated through a benchmark and human annotation protocol. Such a framework would enable interfaces that communicate not just whether a claim is grounded, but how.