Causal Interpretability

Follow this topic Ask about this topic

Momentum

6 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 69

All topics

Right Answer, Wrong Mechanism: Detecting Pernicious Divergence in Causal Interventions
Sep 30, 2026Beiming Liu, Minjie ChenCircuits in Language Models Causal Interpretability
PetroChina Southwest Oil and Gasfield Company, China
Small transformers track Bayesian evidence for latent common causes via a context-invariant mechanism
Sep 28, 2026Amir Mohammadpour, Michael FrankeBayesian Inference Causal Reasoning in Language Models
Department of Linguistics, University of Tübingen
From Concept Alignment to Causal Grounding: An Intervention Test of Chain-of-Thought Faithfulness
Sep 19, 2026Qianli Wang, Yilong Wang, Dennis Wei +5CoT Faithfulness LLM Interpretability
Technische Universität Berlin · German Research Center for Artificial Intelligence (DFKI) · IBM Research +4
Same Outcome, Different Readout: What Does a Steerable Valence Direction in LLMs Represent?
Sep 19, 2026Weihan Li, Xinlei Chen, Yuhan Song +2Language Model Steering LLM Interpretability
The University of Tokyo · Harbin Institute of Technology, Shenzhen · The Hong Kong University of Science and Technology
Splitting the Difference: Interpretable Causal Forests for Treatment Effect Heterogeneity and Bias
Sep 15, 2026Nicolas Alexander Ihlo, Merle BehrCausal Effect Estimation Conditional Average Treatment Effect Estimation
Faculty of Informatics and Data Science University of Regensburg, Germany
Calibrating Interpretability Instruments Before Trusting Their Verdicts
Sep 13, 2026Orion Reblitz-RichardsonLLM Interpretability Causal Interpretability
Distiller Labs
Lagged Coupling: Internal Representations Become Readable Before They Become Causal
Sep 1, 2026Xining XunCausal Interpretability Causal Interventions in Language Models
Tsingjiao Information Science (Beijing) Co., Ltd.
CAS: A Causal Attribution Score for Local and Global Explainable Artificial Intelligence
Aug 12, 2026Michael Georgiades, Charalambia VarnavaFeature Attribution Causal Attribution
Department of Computer Science Neapolis University Pafos Cyprus · CaSToRC The Cyprus Institute Nicosia, Cyprus
Causal Structure is Inducible but Functionally Decoupled: The Routing/Readout Boundary of a Typed Mechanism Library
Aug 12, 2026Xining XunCausal Representation Learning LLM Auditing
Tsingjiao Information Science (Beijing) Co., Ltd.
Visualizing Graph-to-Answer Mechanism Recovery in Materials-Science Hypothesis Generation
Aug 4, 2026Shashwat Sourav, Subhadeep Pal, Markus J. Buehler +4LLM Reasoning with Graphs Scientific Hypothesis Generation
Department of Physics, Washington University in St. Louis · Oak Ridge National Laboratory · Lawrence Berkeley National Laboratory +6
GENESIS: Towards Explainable Causal Discovery
Aug 4, 2026Abhinav Thorat, Ravi Kumar Kolla, Vishak K Bhat +2Causal Discovery Causal Structure Learning
User Engagement Research, Sony Research India
Understanding Online Failure Prediction in Linux Through Complementary Multi-View Explainability
Aug 1, 2026Diogo Dória, João R. CamposFault Detection Causal Interpretability
University of Coimbra, CISUC/LASI Department of Informatics Engineering Coimbra, Portugal
Evidence-Type Competition: When Can Interventional Data Teach Language Models Causal Direction?
Jul 31, 2026Xining XunCausal Discovery Causal Reasoning in Language Models
Tsingjiao Information Science (Beijing) Co., Ltd.
Same Evidence, Different Target: Decoding How Diagnostic Evidence Bears on Causal Questions from Language-Model States
Jul 29, 2026Weiyi Kong, Zhuoran LiCausal Reasoning in Language Models Causal Inference
University of Toronto, Toronto, Canada · The University of Hong Kong, Hong Kong
From Training to Deployment: Post-Hoc Causal Feature Identification via Sensitivity Ratios
Jul 28, 2026Athanasios Vlontzos, Giorgos Papanastasiou, Bernhard Kainz +1Feature Attribution Causal Attribution
Hologen AI London, UK · Mathematics Research Centre Academy of Athens Athens, GR · FAU Erlangen Nuremberg Imperial College London Erlangen, DE & London, UK +1
LAWFUL: Law-Aligned Witness for Faithful Use of Latents
Jul 26, 2026Kevin Chen, Kenneth W. Parker, Anish AroraTransformer Interpretability Mechanistic Interpretability
Computer Science & Engineering, The Ohio State University, Columbus, Ohio, USA · Independent Researcher, Waterford, Virginia, USA
Beyond Sufficiency: Time Series Explanation with Counterfactual Necessity
Jul 23, 2026Hongnan Ma, Yiwei Shi, Mengyue Yang +1Counterfactual Explanations Time Series Classification
School of Computer Science, University of Bristol · School of Engineering Mathematics and Technology, University of Bristol
Are Single-Token Sparse Autoencoder Features Causally Necessary? Layer-Depth and SAE-Family Effects
Jul 22, 2026Seonglae Cho, Zekun Wu, Kleyton Da Costa +3LLM Interpretability Sparse Autoencoders
Holistic AI · University College London
Causal dictionary learning reveals and validates transcription-factor binding features in genomic language models
Jul 21, 2026Sarwan AliLLM Interpretability Genomics
Columbia University Irving Medical Center, New York, USA
Interventional Causal Circuits for Safe Robot Action Testing and Failure Recovery
Jul 16, 2026Naren Vasantakumaar, Tom Schierenbeck, Michael BeetzRobot Failure Detection Robot Failure Recovery
Institute for Artificial Intelligence, University of Bremen
From Geometric Recovery to Causal Validation: A Reproducible Audit of Sparse Autoencoder Features, from Superposition Geometry to Causal Inertness
Jul 13, 2026Mohamed Abdessalem BalFeature Superposition Neural Representation Geometry
Independent Researcher, Algiers, Algeria
Driving the Wrong Way: Leveraging Interpretability in End2End Autonomous Driving Models
Jul 7, 2026Franz Motzkus, Sebastian BernhardEnd-to-End Autonomous Driving Interpretable ML
AUMOVIO, Germany · University of Bamberg, Germany
Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?
Jun 30, 2026Philippe Chlenski, Zachariah Carmichael, Ayush Warikoo +5Faithfulness of Language Model Explanations LLM Interpretability
1Meta.
FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization
Jun 27, 2026Junyan Tan, Yifan Li, Minghao Wang +2Temporal GNNs Causal Interpretability
Department of Computer Science Zhejiang University
Global Explanations for Multivariate Time Series Forecasting Models via $K$ -Order Markov Approximations
Jun 25, 2026Amadeo TunyiMultivariate Time Series Forecasting Surrogate Modeling
XITASO GmbH Austraße 35, 86153 Augsburg, Germany
Beyond Shapley: Efficient Computation of Asymmetric Shapley Values
Jun 23, 2026Ezequiel Companeetz, Santiago Cifuentes, Sergio AbriolaCooperative Game Theory Feature Attribution
Departamento de Computación, Facultad de Ciencias Exactas y Naturales, UBA · Instituto de Ciencias de la Computación (ICC), CONICET
Local Causal Attribution of Chain-of-Thought Reasoning
Jun 20, 2026Dennis Wei, Yannis Belkhiter, Erik Miehling +1LLM Interpretability Causal Attribution
1IBM Research.
CIExplainer++: Generating Causal and Interpretable Explanations for Graph Neural Networks
Jun 17, 2026Francisco Caldas, Sahil Satish Kumar, Ruben Belo +1Causal Interpretability GNN Explainability
NOVA LINCS, NOVA School of Science and Technology, Lisbon, Portugal
The Representational Limit of Scalar Interactions: An Interventional Decomposition
Jun 17, 2026Potito Aghilar, Sabino Roccotelli, Stanislao Fidanza +3Feature Interaction Modeling Representation Disentanglement
Polytechnic University of Bari · University of Bari Aldo Moro
Feature Attribution in Directed Acyclic Graphs Using Edge Intervention
Jun 13, 2026Qiheng Sun, Junxu Liu, Xiaokai Mao +4Feature Attribution Shapley Value Attribution
The Hong Kong Polytechnic University, Hong Kong, China · Zhejiang University, Hangzhou, China · University of Illinois Urbana-Champaign, Urbana, USA
CADET: Physics-Grounded Causal Auditing and Training-Free Deconfounding of End-to-End Driving Planners
Jun 12, 2026Zikun GuoEnd-to-End Autonomous Driving Autonomous Driving Planning
School of Intelligent Driving, Xinwei Institute of Artificial Intelligence · Geely Central Research Institute
When Does Routing Become Interpretable? Causal Probes on Block Attention Residuals
Jun 11, 2026Aydin JavadovTransformer Interpretability Attention Residual Connections
ETH Zurich
Observable Patterns Are Not Explanations: A Causal-Geometric Analysis of Latent Reasoning Models
Jun 10, 2026Darpan Aswal, Thomas Palmeira Ferraz, Yongxin Zhou +1Continuous Latent Reasoning Causal Interpretability
Université Grenoble Alpes, CNRS, Grenoble INP, LIG · Université Paris-Saclay · NAVER LABS Europe
From Observation to Intervention: A Causal Audit of Expert Importance in Mixture-of-Experts Models
Jun 9, 2026Leonard Engmann, Christian Medeiros Adriano, Holger GieseMixture-of-Experts Pruning LLM Auditing
Hasso Plattner Institute, University of Potsdam, Germany.
Many Circuits, One Mechanism: Input Variation and Evaluation Granularity in Circuit Discovery
Jun 4, 2026Alireza Bayat Makou, Jingcheng Niu, Subhabrata Dutta +1Circuit Discovery Mechanistic Interpretability
UKP Lab, Technical University of Darmstadt · National Research Center for Applied Cybersecurity ATHENE
LLM Explainability with Counterfactual Chains and Causal Graphs
Jun 4, 2026Nirit Nussbaum-Hoffer, Nitay Calderon, Liat Ein-Dor +1LLM Reasoning with Graphs LLM Interpretability
Faculty of Data and Decision Sciences, Technion · IBM Research
EML-CD: Causal Mechanism Recovery via EML Symbolic Trees in Structure Learning
Jun 4, 2026Sota AsanumaStructural Causal Models Causal Structure Learning
SoftBank Corp.
Output Type Before Quality: A Standards-Derived XAI Admissibility Rubric for Autonomous-Driving Safety
Jun 3, 2026Abhinaw Priyadershi, Mandar Pitale, Jelena Frtunikj +1AI Assurance Autonomous Driving Safety Evaluation
NVIDIA Corporation, Santa Clara, CA, USA · NVIDIA GmbH, Munich, Germany
When and How Long? The Readout-Mediator Angle in Temporal Reasoning
May 27, 2026Shreyas Fadnavis, Praitayini Kanakaraj, Felix WyssTemporal Reasoning in Language Models Temporal Reasoning
Bioscope AI
MechRL: Reinforcement Learning Agents Perform Circuit Discovery for Mechanistic Interpretability
May 25, 2026Barsat KhadkaTransformer Interpretability Circuit Discovery
School of Computing Sciences and Engineering The University of Southern Mississippi
Causal Tongue-Tie: LLMs Can Encode Causal Direction, But Their Yes/No Outputs Fail to Express
May 25, 2026Ziyi Ding, Xiao-Ping ZhangLLM Evaluation Causal Reasoning in Language Models
Tsinghua Shenzhen International Graduate School, Tsinghua University Shenzhen, China
Represented Is Not Computed: A Causal Test of Candidate Algorithmic Intermediates in a Transformer
May 21, 2026Ishita Darade, Sushrut ThoratNumerical Reasoning in Language Models Transformer Interpretability
MKSSS’s Cummins College of Engineering for Women, Pune 411052, India · Institute of Cognitive Science, Osnabrück University, Osnabrück 49069, Germany
From Correlation to Cause: A Five-Stage Methodology for Feature Analysis in Transformer Language Models
May 21, 2026Caleb MunigetyTransformer Interpretability Causal Reasoning in Language Models
Independent Researcher
Geometry-Adaptive Explainer for Faithful Dictionary-Based Interpretability under Distribution Shift
May 21, 2026Sungjun Lim, Heedong Kim, Andrew Lee +1Distribution Shift Robustness OOD Generalization
aYonsei University · bHarvard University
From Circuit Evidence to Mechanistic Theory: An Inductive Logic Approach
May 20, 2026Nura Aljaafari, Danilo S. Carvalho, Andre FreitasCircuit Discovery Mechanistic Interpretability
Department of Computer Science, University of Manchester, United Kingdom · CRUK National Biomarker Centre, University of Manchester, United Kingdom · Idiap Research Institute, Switzerland
Rethinking Visual Attribution for Chest X-ray Reasoning in Large Vision Language Models
May 19, 2026Guangzhi Xiong, Qiao Jin, Sanchit Sinha +2Medical VQA VLM Interpretability
University of Virginia · National Institutes of Health
Learning Quantifiable Visual Explanations Without Ground-Truth
May 18, 2026Amritpal Singh, Andrey Barsky, Mohamed Ali Souibgui +2Perturbation-Based Feature Attribution Causal Interpretability
Computer Vision Center, Barcelona, Spain · Autonomous University of Barcelona, Spain
OCCAM: Open-set Causal Concept explAnation and Ontology induction for black-box vision Models
May 18, 2026Chiara Maria Russo, Simone Carnemolla, Simone Palazzo +3Explainable Artificial Intelligence Concept-Based Explanations
University of Catania Catania, Italy
Causal Explanations from the Geometric Properties of ReLU Neural Networks
May 11, 2026Hector Woods, Philippa Ryan, Rob AlexanderNeural Network Interpretability Causal Explanation
Department of Computer Science University of York York, United Kingdom
Causal Dimensionality of Transformer Representations: Measurement, Scaling, and Layer Structure
May 9, 2026Nilesh Sarkar, Dawar Jyoti DekaTransformer Interpretability Sparse Autoencoders
Erdős AI Lab
Semiparametric Efficient Test for Interpretable Distributional Treatment Effects
May 8, 2026Houssam Zenati, Arthur GrettonSemiparametric Inference Causal Effect Estimation
1Gatsby Computational Neuroscience Unit, University College London
Position: Mechanistic Interpretability Must Disclose Identification Assumptions for Causal Claims
May 8, 2026Zezheng Lin, Fengming LiuMechanistic Interpretability Causal Identifiability
Preprint.
Radiologist-Guided Causal Concept Bottleneck Models for Chest X-Ray Interpretation
May 8, 2026Amy Rafferty, Rishi Ramaesh, Ajitha RajanChest X-Ray Classification Concept Bottleneck Models
University of Edinburgh, UK · NHS Lothian, UK
PLOT: Progressive Localization via Optimal Transport in Neural Causal Abstraction
May 7, 2026Jonathn Chang, Arya Datla, Ziv GoldfeldMechanistic Interpretability Causal Abstraction
Cornell University
Concept-Based Abductive and Contrastive Explanations for Behaviors of Vision Models
May 7, 2026Ronaldo Canizales, Divya Gopinath, Corina Păsăreanu +1Abductive Reasoning Concept-Based Explanations
Colorado State University, Fort Collins CO, USA · KBR Inc., NASA Ames, Moffett Field CA, USA · Carnegie Mellon University, Pittsburgh PA, USA
Where Reliability Lives in Vision-Language Models: A Mechanistic Study of Attention, Hidden States, and Causal Circuits
May 5, 2026Logan Mann, Ajit Saravanan, Ishan Dave +4VLM Evaluation VLM Interpretability
UC Santa Barbara · UC Berkeley · NVIDIA +2
Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction
May 4, 2026Li Puyin, Jiyuan Tan, Ahmad Jabbar +2Mechanistic Interpretability Causal Abstraction
♢Stanford University · ♠Goodfire
ISAAC: Auditing Causal Reasoning in Deep Models for Drug-Target Interaction
May 3, 2026Barbara Tarantino, Sun Kim, Yijingxiu Lu +1Reasoning Evaluation Causal Interpretability
Department of Economics and Management, University of Pavia, Pavia, Italy · Department of Computer Science and Engineering, Seoul National University, Gwanak-gu 08826, Korea.
Embodied Interpretability: Linking Causal Understanding to Generalization in Vision-Language-Action Models
May 1, 2026Hanxin Zhang, Mingshuo Xu, Abdulqader Dhafer +3VLM Robustness VLM Interpretability
1DANiLab, University of Leicester, UK · School of Computing and Mathematical Sciences, University of Leicester, UK · School of Metallurgy and Materials, University of Birmingham, UK
Causal Learning with Neural Assemblies
Apr 29, 2026Evangelia Kopadi, Dimitris KallesNeural Processes Causal Discovery
Hellenic Open University Greece

Sep 30, 2026, cs.LGJ/K move · Enter open · S save

Right Answer, Wrong Mechanism: Detecting Pernicious Divergence in Causal Interventions

Beiming Liu, Minjie Chen

PetroChina Southwest Oil and Gasfield Company, China

Circuits in Language Models Causal Interpretability Causal Interventions in Language Models Activation Patching

Open paper AskPDF arXiv

Causal interventions such as activation patching and distributed alignment search (DAS) are the main tool for making mechanistic claims about neural networks. Recent work showed that these interventions routinely push representations off the model's natural distribution, and that such divergence is sometimes harmless and sometimes pernicious: it can recruit pathways the model never uses on natural inputs, so that an intervention produces the expected answer through the wrong mechanism. No method currently tells the two cases apart. We make this question testable by planting hidden pathways inside pretrained language models; the pathways are silent on every benchmark prompt by construction, so which interventions depend on them is known exactly. Across 72 configurations and 100,800 interventions on GPT-2 small, we find three things. (i) Nearest-neighbour and local-PCA distances at the intervention site, as used in prior work, score below chance (AUROC 0.35-0.47) at picking out interventions that give the right answer through a planted pathway. (ii) Hidden-Pathway Contribution (HPC), a label-free test that clamps downstream units to the regime of natural runs with the same output and measures how much of the decision disappears, flags pathway-dominated interventions with AUROC >= 0.99 when the pathway shows up as unit-level out-of-regime activity, but fails when every unit stays within its natural range, which we identify as the open problem. (iii) Optimised interventions actively seek hidden pathways: on a gender task, DAS routes 90-95% of its successes through planted pathways for three of four families, and a downstream on-manifold penalty cuts this share to under 5% at a cost of 6-11 points of success rate. In unmodified GPT-2, successful interventions show almost no unit-level out-of-regime reliance.

Momentum

Related topics

Latest papers 69