Mechanistic Interpretability papers, 2023–2026

Key findings

• Mechanistic Interpretability went from 17 papers in 2023 to 138 in 2026, roughly 8.1x in three years.
• ICML publishes the most Mechanistic Interpretability work: 119 papers from 2023 to 2026, 38% of the four-venue total.
• The fastest-growing overlap is with Deep learning · large language models: papers tagged with both rose from 1 in 2023 to 23 in 2026.
• Among the 5 topics under Interpretability, it ranks #2 by 2026 paper count; the fastest grower in the group, Feature Attribution, is up 1.9x versus 8.1x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026A Mechanistic Interpretability Study of an Astronomical Foundation Model
• ICML 2026Contribution Weights: A Geometrical Analysis of Self-Attention Transformers
• ICLR 2026Circuit Insights: Towards Interpretability Beyond Activations
• CVPR 2026CIGMA Causal Information-Gain Mechanistic Attribution of Attention Heads in Vision
• NeurIPS 2025Beyond Components: Singular Vector-Based Interpretability of Transformer Circuits
• ICML 2025Algorithm Development in Neural Networks: Insights from the Streaming Parity Task
• ICLR 2025Arithmetic Without Algorithms: Language Models Solve Math with a Bag of Heuristics
• CVPR 2025Dissecting and Mitigating Diffusion Bias via Mechanistic Interpretability
• NeurIPS 2024InterpBench: Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques
• ICML 2024Mechanistic Neural Networks for Scientific Machine Learning
• ICLR 2024Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking
• CVPR 2024SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology
• NeurIPS 2023A Logic for Expressing Log-Precision Transformers
• ICML 2023Mechanistic Mode Connectivity
• ICLR 2023Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 Small
• NeurIPS 2026Discovering Mechanistic Models of Neural Activity: System Identification in an in Silico Zebrafish
• ICML 2026CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features
• ICLR 2026CLUE: Conflict-guided Localization for LLM Unlearning Framework
• CVPR 2026From Weights to Concepts Data-Free Interpretability of CLIP via Singular
• NeurIPS 2025Bigram Subnetworks: Mapping to Next Tokens in Transformer Language Models

All Mechanistic Interpretability papers

102 papers, page 1 of 3

1. ICLR 2026AbsTopK: Rethinking Sparse Autoencoders For Bidirectional Features
2. ICLR 2026Activation Steering with a Feedback Controller
3. ICLR 2026ActivationReasoning: Logical Reasoning in Latent Activation Spaces
4. ICLR 2026ASGuard: Activation-Scaling Guard to Mitigate Targeted Jailbreaking Attack
5. ICLR 2026Attributing Response to Context: A Jensen–Shannon Divergence Driven Mechanistic Study of Context Attribution in Retrieval-Augmented Generation
6. ICLR 2026Attribution-Guided Decoding
7. ICLR 2026Characterizing Pattern Matching and Its Limits on Compositional Task Structures
8. ICLR 2026Circuit Insights: Towards Interpretability Beyond Activations
9. ICLR 2026CLUE: Conflict-guided Localization for LLM Unlearning Framework
10. ICLR 2026Correlations in the Data Lead to Semantically Rich Feature Geometry Under Superposition
11. ICLR 2026Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning
12. ICLR 2026Discovering and Steering Interpretable Concepts in Large Generative Music Models
13. ICLR 2026Does Higher Interpretability Imply Better Utility? A Pairwise Analysis on Sparse Autoencoders
14. ICLR 2026Dyslexify: A Mechanistic Defense Against Typographic Attacks in CLIP
15. ICLR 2026Eliciting Numerical Predictive Distributions of LLMs Without Auto-Regression
16. ICLR 2026Evolution of Concepts in Language Model Pre-Training
17. ICLR 2026Formal Mechanistic Interpretability: Automated Circuit Discovery with Provable Guarantees
18. ICLR 2026Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition
19. ICLR 2026Hedonic Neurons: A Mechanistic Mapping of Latent Coalitions in Transformer MLPs
20. ICLR 2026How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic Interpretability
21. ICLR 2026Internal Planning in Language Models: Characterizing Horizon and Branch Awareness
22. ICLR 2026Language Models Use Lookbacks to Track Beliefs
23. ICLR 2026Latent Concept Disentanglement in Transformer-based Language Models
24. ICLR 2026Latent Planning Emerges with Scale
25. ICLR 2026Learning Concept Bottleneck Models from Mechanistic Explanations
26. ICLR 2026Localizing Task Recognition and Task Learning in In-Context Learning via Attention Head Analysis
27. ICLR 2026Mechanism of Task-oriented Information Removal in In-context Learning
28. ICLR 2026Mechanistic Independence: A Principle for Identifiable Disentangled Representations
29. ICLR 2026MICLIP: Learning to Interpret Representation in Vision Models
30. ICLR 2026Mixing Mechanisms: How Language Models Retrieve Bound Entities In-Context
31. ICLR 2026Narrow Finetuning Leaves Clearly Readable Traces in the Activation Differences
32. ICLR 2026On the Predictive Power of Representation Dispersion in Language Models
33. ICLR 2026Path Channels and Plan Extension Kernels: a Mechanistic Description of Planning in a Sokoban RNN
34. ICLR 2026Provably Tracking Equivalent Mechanistic Interpretations Across Neural Networks
35. ICLR 2026Reforming the Mechanism: Editing Reasoning Patterns in LLMs with Circuit Reshaping
36. ICLR 2026Rethinking Layer Relevance in Large Language Models Beyond Cosine Similarity
37. ICLR 2026Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs
38. ICLR 2026Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insights
39. ICLR 2026The Deleuzian Representation Hypothesis
40. ICLR 2026The Geometry and Topology of Circuits: the Manifolds of Modular Addition
41. ICLR 2026The Price of Amortized inference in Sparse Autoencoders
42. ICLR 2026Thought Branches: Interpreting LLM Reasoning Requires Resampling
43. ICLR 2026Towards Understanding the Nature of Attention with Low-Rank Sparse Decomposition
44. ICLR 2026Transformers Don’t Need LayerNorm at Inference Time: Scaling LayerNorm Removal to GPT-2 XL and Implications for Mechanistic Interpretability
45. ICLR 2026Verifying Chain-of-Thought Reasoning via Its Computational Graph
46. ICLR 2026When Thinking Backfires: Mechanistic Insights into Reason-induced Misalignment
47. ICLR 2025Adaptive Transformer Programs: Bridging the Gap Between Performance and Interpretability in Transformers
48. ICLR 2025Arithmetic Without Algorithms: Language Models Solve Math with a Bag of Heuristics
49. ICLR 2025Attention as a Hypernetwork
50. ICLR 2025Bilinear MLPs enable weight-based mechanistic interpretability

FAQ

How many Mechanistic Interpretability papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

311 in total: 17 in 2023, 49 in 2024, 107 in 2025, 138 in 2026.

How many Mechanistic Interpretability papers are at NeurIPS 2026?

NeurIPS 2026 has 16 Mechanistic Interpretability papers, versus 34 the year before.

How many Mechanistic Interpretability papers are at ICLR 2026?

ICLR 2026 has 46 Mechanistic Interpretability papers, versus 38 the year before.

How many Mechanistic Interpretability papers are at ICML 2026?

ICML 2026 has 74 Mechanistic Interpretability papers, versus 33 the year before.

How many Mechanistic Interpretability papers are at CVPR 2026?

CVPR 2026 has 2 Mechanistic Interpretability papers, versus 2 the year before.

Which conference publishes the most Mechanistic Interpretability papers?

ICML, with 119 papers from 2023 to 2026.