Schedule

Day 1: 15 August 2026
Building GW2, room B3009

9:00 Welcome

9:10 Keynote talk: Vera Schmitt, "From Model Behaviour to Human Understanding: The Case for Actionable Explainable AI"

9:40 Paper session: Interpretable Machine Learning

• Part-based Quantitative Analysis for Heatmaps - Osman Tursun, Sinan Kalkan, Simon Denman, Sridha Sridharan, Clinton Fookes 

• Out-of-Distribution Detection using Counterfactual Distance - Maria Stoica, Francesco Leofante, Alessio Lomuscio

• Semirings Meet Interpretability: A Quantitative Logic for Model Explanations - Alexander Pinto, Marcelo Arenas 

• The Directed Prediction Change - Efficient and Trustworthy Fidelity Assessment for Local Feature Attribution Methods - Kevin Kim Iselborn, David Dembinsky, Adriano Lucieri, Andreas Dengel 

10:40 Coffee break

11:00 Paper session: Explainable Sequential Decision Making

• Command-Space Counterfactual Explanations for Pareto-Conditioned Reinforcement Learning - Joanikij Chulev, Hendrik Baier 

• Toward Template-Free Explainability for Monte Carlo Tree Search - Siqi Lu, Mirsaleh Bahavarnia, Hiba Baroud, Yixuan Zhang, Hemant Purohit, Ayan Mukhopadhyay 

• Assessing Automatic Concept Extraction for Reinforcement Learning Policies - Chengpeng Hu, Yingqian Zhang, Hendrik Baier 

• Evaluating RL Explainability Methods by How Much They Help Fix Bugs in Agents - Ram Rachum, Yotam Amitai, Bálint Gyevnár, Reuth Mirsky, Cameron Allen 

12:00 Lunch break (lunch not provided by IJCAI)

13:00 Paper session: Human-Centered XAI

• Hu-CEMNIST: A Benchmark Dataset of Human-Generated Counterfactual Explanations for MNIST - Ulrike Kuhl, André Artelt 

• Acceptance and Perception of AI: The Roles of Model Confidence, Correctness, and Counterfactual Explanations Formats - Xinyue Dai, Georgios Fragkoulidis, Guro Braut, Paul Liston 

• Addressing the Selection Problem in Explainable AI - Claire Vlases, Katelyn Morrison 

• Beyond Output Confidence - A Mechanistic Interpretability Analysis of Sycophantic Vulnerability in Large Language Models - Harleen Kaur Kaur Bagga, Justin Shenk 

14:00 Coffee break

14:20 Paper session

• Global Explanations for Multivariate Time Series Forecasting Models via K-Order Markov Approximations - Amadeo Tunyi 

• The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations - Amadeo Tunyi  

• FlagGAM: Rule-Based Generalized Additive Modeling for Explainable Tabular Prediction - Zijie Zhao, Roy E. Welsch 

• When Superpixels Fail on Documents: A Study of Segmentation for LIME Explanations - Quentin Telnoff, Emanuela Boros, Mickael Coustaty, Robin Jarry, Fabrice Crohas, Antoine Doucet 

15:20 Paper session

• Improving Human Oversight of AI Systems With Expert Feedback Using Interactive, Contrastive Explanations - Zhengyu Su, Mark T Keane, Eoin Delaney 

• Do Neural Networks Attend to Expert-Recognized Morphological Characters? An XAI Case Study in Small-Data Zooplankton Classification - Valeria Kirova, Petr Garibian1, Vladimir Valeyev, Anna Kasatkina, Valentin Toptunov, Egor Belianin, Vitaliia Rubleva 

• Towards an Argumentative Foundation for Evaluative AI - Xiang Yin, Tim Miller, Nico Potyka, Antonio Rago, Francesca Toni 

• Counterfactual Explanations Under Concept Drift - Marcin Kostrzewa, Jerzy Stefanowski, Maciej Zięba 

• On the Reliability of Post-Hoc Attributions Under Adversarial Training - Maximilian Wendlinger, Paul-Andrei Sava, Wei Herng Choong, Ching-Yu Kao, Daniel Kowatsch 

16:20 End of day 1

Day 2: 16 August 2026
Building GW2, room B3009

9:00 Welcome

9:10 Keynote talk: Francesca Toni, "Explainable AI for Contestability"

9:40 Paper session: Interpretable Machine Learning

• Why Textual Rule Explanations Drift: A Mechanism-Level Diagnosis of Stability Failures in RL - Yulong Wu 

• Does Target Class Matter? A Reproducibility Study of Plausible Counterfactual Explanations for Image Classification - Jasmin Kareem, Annelot Bosman, Meike Nauta, Martijn Willemsen, Jan van Rijn, Maarten de Rijke 

• From Decoding to Explanation: Selecting and Validating Brain Regions in Task-fMRI - Valeria Kirova, Ivan Butakov, Alexander Semenenko, Alexey Frolov 

• Evaluating Explanation-Driven Vision–Language Reasoning via Generation Order Interventions - Siting Liang, Luca Rippe, Omar Adjali, Daniel Sonntag 

10:40 Coffee break

11:00 Paper session: Explainable Sequential Decision Making

• CSTA++: Unifying State, Action Temporal Abstractions and Causal Information for Contrastive Policy Explanations - Xiaowei Liu, Yining Yuan, Weiru Liu 

• BXRL: Behavior-Explainable Reinforcement Learning - Ram Rachum, Yotam Amitai, Yonatan Nakar, Reuth Mirsky, Cameron Allen 

• Forecasting-Conditioned Reinforcement Learning: Anticipatable Policies via Multi-Step Self-Forecasting - Jeremias Gerner, Klaus Bogenberger, Stefanie Schmidtner 

• Sequential Decision-Making with Explanatory Actions - Hendrik Baier, Roya Daneshi, Sarath Sreedharan 

12:00 Lunch break (lunch not provided by IJCAI)

13:00 Paper session: Human-Centered XAI

• Explaining Risk in High Stakes Domains - Claudia V. Goldman, Assaf Gozlan 

• T̶w̶o̶ ̶B̶l̶a̶c̶k̶ ̶B̶o̶x̶e̶s̶,̶ ̶O̶n̶e̶ ̶S̶o̶l̶v̶e̶r̶:̶ ̶E̶n̶c̶o̶d̶e̶r̶ ̶P̶r̶o̶b̶i̶n̶g̶ ̶a̶n̶d̶ ̶D̶e̶c̶o̶d̶e̶r̶ ̶A̶t̶t̶r̶i̶b̶u̶t̶i̶o̶n̶ ̶f̶o̶r̶ ̶N̶e̶u̶r̶a̶l̶ ̶M̶u̶l̶t̶i̶-̶A̶t̶t̶r̶i̶b̶u̶t̶e̶ ̶V̶R̶P̶ ̶u̶n̶d̶e̶r̶ ̶H̶a̶r̶d̶-̶M̶a̶s̶k̶ ̶a̶n̶d̶ ̶R̶e̶c̶o̶u̶r̶s̶e̶ ̶D̶e̶c̶o̶d̶e̶r̶s̶  presentation cancelled

• Counterfactuals by Design: Residual Decomposition and Back-to-Normality Editing for Time Series - Szymon Bobek, Syed Muhammad Hamza Zaidi, Myra Spiliopoulou, Grzegorz J. Nalepa 

• Faithfulness Without Anchors: A Position Paper on the Structural Misalignments of Explainable AI Faithfulness Metrics for Feature Importance - Marco Zullich, Emily Schiller, Ivan Gentile, David Dembinsky, Adriano Lucieri, Kilian Göller, Steffen Seitz 

13:50 Coffee break

14:10 Paper session: Interpretable Machine Learning

• Beyond Heatmaps: Unsupervised Concept-Graph Reasoning for Interpretable Visual Explanation - Md Mohasin Hossain, Anar Amirli, Robert Andreas Leist, Md Abdul Kadir, Daniel Sonntag 

• Radiologist-Guided Causal Concept Bottleneck Models for Chest X-Ray Interpretation - Amy Rafferty, Rishi Ramaesh, Ajitha Rajan 

• Do Linear Probes Generalize Better in Persona Coordinates?   presentation cancelled

• CRISP - a Methodological Framework for Complexity-Guided Visual Hidden-Layer xAI - Simon Geerkens, Christian Sieberichs, Alexander Braun, Visvanathan Ramesh, Thomas Waschulzik 

• Brain–Model Alignment as a Probe for Subliminal Learning  presentation cancelled

15:00 Poster session

16:00 Fishbowl and closing

17:00 End of day 2