Making Sense of LLM Decisions: A Prototype-based Framework for Explainable Classification

Bowen Wei; Mehrdad Fazli; Ziwei Zhu

AAAI 2026technical0 citations

Making Sense of LLM Decisions: A Prototype-based Framework for Explainable Classification

Abstract

Large language models have demonstrated impressive performance on natural language tasks, but their decision-making processes remain opaque. Existing explanation methods either suffer from limited faithfulness to the model

BibTeX

@inproceedings{aaai2026_makingsenseofllm,
  title = {Making Sense of LLM Decisions: A Prototype-based Framework for Explainable Classification},
  author = {Bowen Wei and Mehrdad Fazli and Ziwei Zhu},
  booktitle = {AAAI 2026},
  year = {2026}
}