Focus-LIME: Surgical Interpretation of Long-Context Large Language Models via Proxy-Based Neighborhood Selection

Junhao Liu; Haonan Yu; Zhenyu Yan; Xin Zhang

IJCAI 20260 citations

Focus-LIME: Surgical Interpretation of Long-Context Large Language Models via Proxy-Based Neighborhood Selection

Junhao Liu, Haonan Yu, Zhenyu Yan, Xin Zhang

Abstract

As Large Language Models (LLMs) scale to handle massive context windows, achieving surgical feature-level interpretation is essential for high-stakes tasks like legal auditing and code debugging. However, existing local model-agnostic explanation methods face a critical dilemma in these scenarios: feature-based methods suffer from attribution dilution due to high feature dimensionality, which prevents them from providing faithful explanations. In this paper, we propose Focus-LIME, a coarse-to-fine framework designed to restore the tractability of surgical interpretation. Focus-LIME utilizes a proxy model to curate the perturbation neighborhood, allowing the target model to perform fine-grained attribution exclusively within the optimized context. Empirical evaluations on long-context benchmarks demonstrate that our method makes surgical explanations practical and provides faithful explanations to users.

Machine Learning: Explainable/Interpretable machine learningNatural Language Processing: Interpretability and analysis of models for NLP

BibTeX

@inproceedings{ijcai2026_focuslimesurgica,
  title = {Focus-LIME: Surgical Interpretation of Long-Context Large Language Models via Proxy-Based Neighborhood Selection},
  author = {Junhao Liu and Haonan Yu and Zhenyu Yan and Xin Zhang},
  booktitle = {IJCAI 2026},
  year = {2026}
}