2026

LINE: LLM-based Iterative Neuron Explanation

2025

When Interpretability Hurts Information Retrieval