# Concept Bottleneck Large Language Models

arXiv: https://arxiv.org/abs/2412.07992 · HTML: https://mlsanity.com/paper/2412.07992/

## Key facts

- Cited by 43 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 34 references that are in the map.
- First posted on arXiv 11 Dec 2024.

## Builds on (most cited first, 16 of 16)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification](https://mlsanity.com/paper/2211.11158/) (2022)
- [Post-hoc Concept Bottleneck Models](https://mlsanity.com/paper/2205.15480/) (2022)
- [Label-Free Concept Bottleneck Models](https://mlsanity.com/paper/2304.06129/) (2023)
- [CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networks](https://mlsanity.com/paper/2204.10965/) (2022)
- [Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery](https://mlsanity.com/paper/2407.14499/) (2024)
- [VLG-CBM: Training Concept Bottleneck Models with Vision-Language Guidance](https://mlsanity.com/paper/2408.01432/) (2024)
- [Interpreting Pretrained Language Models via Concept Bottlenecks](https://mlsanity.com/paper/2311.05014/) (2023)
- [Energy-Based Concept Bottleneck Models: Unifying Prediction, Concept Intervention, and Probabilistic Interpretations](https://mlsanity.com/paper/2401.14142/) (2024)
- [Classification with Conceptual Safeguards](https://mlsanity.com/paper/2411.04342/) (2024)

## Built on by (most cited first, 30 of 43)

- [Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models](https://mlsanity.com/paper/2601.14004/) (2026)
- [Sparse Autoencoders for Hypothesis Generation](https://mlsanity.com/paper/2502.04382/) (2025)
- [LLM Agents Already Know When to Call Tools -- Even Without Reasoning](https://mlsanity.com/paper/2605.09252/) (2026)
- [A Comprehensive Survey on Self-Interpretable Neural Networks](https://mlsanity.com/paper/2501.15638/) (2025)
- [ThinkEdit: Interpretable Weight Editing to Mitigate Overly Short Thinking in Reasoning Models](https://mlsanity.com/paper/2503.22048/) (2025)
- [Interpretability without actionability: mechanistic methods cannot correct language model errors despite near-perfect internal representations](https://mlsanity.com/paper/2603.18353/) (2026)
- [Interpretable Generative Models through Post-hoc Concept Bottlenecks](https://mlsanity.com/paper/2503.19377/) (2025)
- [Object Centric Concept Bottlenecks](https://mlsanity.com/paper/2505.24492/) (2025)
- [Interpretable and Steerable Concept Bottleneck Sparse Autoencoders](https://mlsanity.com/paper/2512.10805/) (2025)
- [An Analysis of Concept Bottleneck Models: Measuring, Understanding, and Mitigating the Impact of Noisy Annotations](https://mlsanity.com/paper/2505.16705/) (2025)
- [ReflCtrl: Controlling LLM Reflection Efficiently via Representation Engineering](https://mlsanity.com/paper/2512.13979/) (2025)
- [Interpreting and Steering State-Space Models via Activation Subspace Bottlenecks](https://mlsanity.com/paper/2602.22719/) (2026)
- [Cross-Lingual LLM-Judge Transfer via Evaluation Decomposition](https://mlsanity.com/paper/2603.18557/) (2026)
- [Explainability of Large Language Models: Opportunities and Challenges toward Generating Trustworthy Explanations](https://mlsanity.com/paper/2510.17256/) (2025)
- [Concept-Guided Backdoor Attack on Vision Language Models](https://mlsanity.com/paper/2512.00713/) (2025)
- [Steer2Edit: From Activation Steering to Component-Level Editing](https://mlsanity.com/paper/2602.09870/) (2026)
- [Mitigating Translationese Bias in Multilingual LLM-as-a-Judge via Disentangled Information Bottleneck](https://mlsanity.com/paper/2603.10351/) (2026)
- [Understanding Fixed Predictions via Confined Regions](https://mlsanity.com/paper/2502.16380/) (2025)
- [Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification](https://mlsanity.com/paper/2509.20899/) (2025)
- [Hierarchical Concept Embedding & Pursuit for Interpretable Image Classification](https://mlsanity.com/paper/2602.11448/) (2026)
- [UrbanAlign: Post-hoc Semantic Calibration for VLM-Human Preference Alignment](https://mlsanity.com/paper/2602.19442/) (2026)
- [Revealing Combinatorial Reasoning of GNNs via Graph Concept Bottleneck Layer](https://mlsanity.com/paper/2603.02025/) (2026)
- [SCoCCA: Multi-modal Sparse Concept Decomposition via Canonical Correlation Analysis](https://mlsanity.com/paper/2603.13884/) (2026)
- [Which Concepts to Forget and How to Refuse? Decomposing Concepts for Continual Unlearning in Large Vision-Language Models](https://mlsanity.com/paper/2603.21484/) (2026)
- [Generative Interpretability via Scalable Neuro-Symbolic Models](https://mlsanity.com/paper/2609.13529/) (2026)
- [Wanting to Be Understood Explains the Meta-Problem of Consciousness](https://mlsanity.com/paper/2506.12086/) (2025)
- [Chat-CBM: Towards Interactive Concept Bottleneck Models with Frozen Large Language Models](https://mlsanity.com/paper/2509.17522/) (2025)
- [Explaining Fine Tuned LLMs via Counterfactuals A Knowledge Graph Driven Framework](https://mlsanity.com/paper/2509.21241/) (2025)
- [ReFIne: A Framework for Trustworthy Large Reasoning Models with Reliability, Faithfulness, and Interpretability](https://mlsanity.com/paper/2510.09062/) (2025)
- [From Black-box to Causal-box: Towards Building More Interpretable Models](https://mlsanity.com/paper/2510.21998/) (2025)

## Abstract (arXiv)

We introduce Concept Bottleneck Large Language Models (CB-LLMs), a novel framework for building inherently interpretable Large Language Models (LLMs). In contrast to traditional black-box LLMs that rely on limited post-hoc interpretations, CB-LLMs integrate intrinsic interpretability directly into the LLMs -- allowing accurate explanations with scalability and transparency. We build CB-LLMs for two essential NLP tasks: text classification and text generation. In text classification, CB-LLMs is competitive with, and at times outperforms, traditional black-box models while providing explicit and interpretable reasoning. For the more challenging task of text generation, interpretable neurons in CB-LLMs enable precise concept detection, controlled generation, and safer outputs. The embedded interpretability empowers users to transparently identify harmful content, steer model behavior, and unlearn undesired concepts -- significantly enhancing the safety, reliability, and trustworthiness of LLMs, which are critical capabilities notably absent in existing models. Our code is available at https://github.com/Trustworthy-ML-Lab/CB-LLMs.