mlsanityData to 14 Oct 2026

Learning to Explain: Prototype-Based Surrogate Models for LLM Classification

arXiv 2505.18970 ↗First posted 25 May 2025cs.CLcited by 3

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) have demonstrated impressive performance on natural language tasks, but their decision-making processes remain largely opaque. Existing explanation methods either suffer from limited faithfulness to the model's reasoning or produce explanations that humans find difficult to understand. To address these challenges, we propose ProtoSurE, a novel prototype-based surrogate framework that provides faithful and human-understandable explanations for LLMs. ProtoSurE trains an interpretable-by-design surrogate model that aligns with the target LLM while utilizing sentence-level prototypes as human-understandable concepts. Extensive experiments show that ProtoSurE consistently outperforms SOTA explanation methods across diverse LLMs and datasets. Importantly, ProtoSurE demonstrates strong data efficiency, requiring relatively few training examples to achieve good performance, making it practical for real-world applications.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 26 of its 38 references that are in the map. First posted on arXiv 25 May 2025.

Builds on

26 of 38 references.

References · 26

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  4. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  5. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  7. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  8. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  9. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
  10. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,073
  11. 2020MPNet: Masked and Permuted Pre-training for Language Understandingcited by 981
  12. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
14 more
  1. 2023Measuring Faithfulness in Chain-of-Thought Reasoningcited by 703
  2. 2020Transformer Interpretability Beyond Attention Visualizationcited by 627
  3. 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 599
  4. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 530
  5. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  6. 2021Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Modelscited by 463
  7. 2019Is Attention Interpretable?cited by 389
  8. 2017Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictionscited by 234
  9. 2024Are self-explanations from Large Language Models faithful?cited by 101
  10. 2019Interpretable and Steerable Sequence Learning via Prototypescited by 72
  11. 2017Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillationcited by 62
  12. 2021Improving the Faithfulness of Attention-based Explanations with Task-specific Information for Text Classificationcited by 29
  13. 2019Leveraging Latent Features for Local Explanationscited by 21
  14. 2024Advancing Interpretability in Text Classification through Prototype Learningcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Cognitive Alignment in Personality Reasoning: Leveraging Prototype Theory for MBTI Inferencecited by 0
  2. 2026Discovering Implicit Large Language Model Alignment Objectivescited by 0
  3. 2026Structured Prompt Optimization Meets Reinforcement Learning for Global and Local Interpretability over Complex Textcited by 0