DBMay 31

Can we trust LLM Self-Explanations for Entity Resolution?

arXiv:2606.0121011.3
Predicted impact top 26% in DB · last 90 daysOriginality Incremental advance
AI Analysis

For practitioners using LLMs for entity resolution, this work reveals that self-explanations are unreliable and offers a cost-effective alternative that balances trustworthiness and efficiency.

The paper evaluates LLM self-explanations for Entity Resolution and finds them unstable, weakly faithful, and poorly aligned with counterfactual evidence. It proposes a hybrid framework, UNCERTA, that achieves explanation quality comparable to post-hoc methods while reducing computational cost by up to an order of magnitude.

Large Language Models (LLMs) have recently shown strong performance on Entity Resolution (ER). Additionally, akin to their prowess in providing accurate predictions, these models often generate self-explanations alongside their predictions through prompting. While such self-explanations are appealing due to their negligible computational cost, their actual reliability remains largely unexplored. In this paper, we present the first large-scale systematic evaluation of LLM self-explanations for ER, focusing on feature attribution and counterfactual explanations at both the attribute and token levels. Across three LLMs, ten datasets, and multiple prompting strategies, we show that self-explanations are often unstable, weakly faithful, and poorly aligned with counterfactual evidence, revealing a substantial gap between plausibility and causal relevance. We further demonstrate that established post-hoc explanation methods provide significantly higher trustworthiness, but at a prohibitive computational cost when applied to LLMs. To bridge this gap, we introduce \uncerta{}, a hybrid explanation framework that leverages self-explanations as priors to guide post-hoc exploration. \uncerta{} achieves explanation quality comparable to post-hoc methods while reducing cost by up to an order of magnitude.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes