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

arXiv: https://arxiv.org/abs/2602.04607 · HTML: https://mlsanity.com/paper/2602.04607/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 34 references that are in the map.
- First posted on arXiv 4 Feb 2026.

## Builds on (most cited first, 21 of 21)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers](https://mlsanity.com/paper/2105.03011/) (2021)
- [LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens](https://mlsanity.com/paper/2402.13753/) (2024)
- [On the (In)fidelity and Sensitivity for Explanations](https://mlsanity.com/paper/1901.09392/) (2019)
- [LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models](https://mlsanity.com/paper/2310.05736/) (2023)
- [CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review](https://mlsanity.com/paper/2103.06268/) (2021)
- [AttnLRP: Attention-Aware Layer-Wise Relevance Propagation for Transformers](https://mlsanity.com/paper/2402.05602/) (2024)
- [ALIME: Autoencoder Based Approach for Local Interpretability](https://mlsanity.com/paper/1909.02437/) (2019)
- [Explain Any Concept: Segment Anything Meets Concept-Based Explanation](https://mlsanity.com/paper/2305.10289/) (2023)
- [Interpretable-by-Design Text Understanding with Iteratively Generated Concept Bottleneck](https://mlsanity.com/paper/2310.19660/) (2023)
- [GLIME: General, Stable and Local LIME Explanation](https://mlsanity.com/paper/2311.15722/) (2023)
- [Evaluating Large Language Models for Health-Related Text Classification Tasks with Public Social Media Data](https://mlsanity.com/paper/2403.19031/) (2024)
- [Are Your Explanations Reliable? Investigating the Stability of LIME in Explaining Text Classifiers by Marrying XAI and Adversarial Attack](https://mlsanity.com/paper/2305.12351/) (2023)
- [ReX: A Framework for Incorporating Temporal Information in Model-Agnostic Local Explanation Techniques](https://mlsanity.com/paper/2209.03798/) (2022)
- [Towards Explainability in Legal Outcome Prediction Models](https://mlsanity.com/paper/2403.16852/) (2024)
- [Rethinking Saliency Maps: A Cognitive Human Aligned Taxonomy and Evaluation Framework for Explanations](https://mlsanity.com/paper/2511.13081/) (2025)
- [MAnchors: Memorization-Based Acceleration of Anchors via Rule Reuse and Transformation](https://mlsanity.com/paper/2502.11068/) (2025)
- [Revitalizing Black-Box Interpretability: Actionable Interpretability for LLMs via Proxy Models](https://mlsanity.com/paper/2505.12509/) (2025)

## Abstract (arXiv)

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, thus failing to provide 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 practicable and provides faithful explanations to users.