# DISCOVER: Making Vision Networks Interpretable via Competition and Dissection

arXiv: https://arxiv.org/abs/2310.04929 · HTML: https://mlsanity.com/paper/2310.04929/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 33 references that are in the map.
- First posted on arXiv 7 Oct 2023.

## Builds on (most cited first, 23 of 23)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [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)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables](https://mlsanity.com/paper/1611.00712/) (2016)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [Network Dissection: Quantifying Interpretability of Deep Visual Representations](https://mlsanity.com/paper/1704.05796/) (2017)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [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)
- [Natural Language Descriptions of Deep Visual Features](https://mlsanity.com/paper/2201.11114/) (2022)
- [Stochastic Transformer Networks with Linear Competing Units: Application to end-to-end SL Translation](https://mlsanity.com/paper/2109.13318/) (2021)
- [Sparse Linear Concept Discovery Models](https://mlsanity.com/paper/2308.10782/) (2023)
- [Nonparametric Bayesian Deep Networks with Local Competition](https://mlsanity.com/paper/1805.07624/) (2018)
- [Contextual Semantic Interpretability](https://mlsanity.com/paper/2009.08720/) (2020)
- [Local Competition and Stochasticity for Adversarial Robustness in Deep Learning](https://mlsanity.com/paper/2101.01121/) (2021)
- [Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness](https://mlsanity.com/paper/2112.02671/) (2021)
- [Stochastic Deep Networks with Linear Competing Units for Model-Agnostic Meta-Learning](https://mlsanity.com/paper/2208.01573/) (2022)
- [Competing Mutual Information Constraints with Stochastic Competition-based Activations for Learning Diversified Representations](https://mlsanity.com/paper/2201.03624/) (2022)

## Built on by (most cited first, 8 of 8)

- [Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery](https://mlsanity.com/paper/2407.14499/) (2024)
- [Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models](https://mlsanity.com/paper/2504.02821/) (2025)
- [A Survey on Safe Multi-Modal Learning System](https://mlsanity.com/paper/2402.05355/) (2024)
- [LG-CAV: Train Any Concept Activation Vector with Language Guidance](https://mlsanity.com/paper/2410.10308/) (2024)
- [Discovering Influential Neuron Path in Vision Transformers](https://mlsanity.com/paper/2503.09046/) (2025)
- [A Geometric Unification of Concept Learning with Concept Cones](https://mlsanity.com/paper/2512.07355/) (2025)
- [Hierarchical Concept-based Interpretable Models](https://mlsanity.com/paper/2602.23947/) (2026)
- [Concept-Based Mechanistic Interpretability Using Structured Knowledge Graphs](https://mlsanity.com/paper/2507.05810/) (2025)

## Abstract (arXiv)

Modern deep networks are highly complex and their inferential outcome very hard to interpret. This is a serious obstacle to their transparent deployment in safety-critical or bias-aware applications. This work contributes to post-hoc interpretability, and specifically Network Dissection. Our goal is to present a framework that makes it easier to discover the individual functionality of each neuron in a network trained on a vision task; discovery is performed in terms of textual description generation. To achieve this objective, we leverage: (i) recent advances in multimodal vision-text models and (ii) network layers founded upon the novel concept of stochastic local competition between linear units. In this setting, only a small subset of layer neurons are activated for a given input, leading to extremely high activation sparsity (as low as only ≈ 4%). Crucially, our proposed method infers (sparse) neuron activation patterns that enables the neurons to activate/specialize to inputs with specific characteristics, diversifying their individual functionality. This capacity of our method supercharges the potential of dissection processes: human understandable descriptions are generated only for the very few active neurons, thus facilitating the direct investigation of the network's decision process. As we experimentally show, our approach: (i) yields Vision Networks that retain or improve classification performance, and (ii) realizes a principled framework for text-based description and examination of the generated neuronal representations.