Towards Intrinsic Interpretability of Large Language Models:A Survey of Design Principles and Architectures
For researchers and practitioners in explainable AI, this paper provides a structured overview of an emerging field, but it is a literature review without novel empirical results.
This survey systematically reviews intrinsic interpretability methods for LLMs, categorizing them into five design paradigms, and outlines open challenges and future directions.
While Large Language Models (LLMs) have achieved strong performance across many NLP tasks, their opaque internal mechanisms hinder trustworthiness and safe deployment. Existing surveys in explainable AI largely focus on post-hoc explanation methods that interpret trained models through external approximations. In contrast, intrinsic interpretability, which builds transparency directly into model architectures and computations, has recently emerged as a promising alternative. This paper presents a systematic review of the recent advances in intrinsic interpretability for LLMs, categorizing existing approaches into five design paradigms: functional transparency, concept alignment, representational decomposability, explicit modularization, and latent sparsity induction. We further discuss open challenges and outline future research directions in this emerging field. The paper list is available at: https://github.com/PKU-PILLAR-Group/Survey-Intrinsic-Interpretability-of-LLMs.