Programme

September 22nd 2025 - Room D3

11:30 – 11:35 Welcome by the organisers

11:35 – 12:30 Tutorial: Practical Bandits: An Industry Perspective

      Bram van den Akker, Olivier Jeunen, Ben London and Emma Kong.

12:30 – 13:30 Lunch Break

13:30 – 14:30 Keynote: Causal Identification: State of the Art and Challenges

       Negar Kiyavash.

                         14:30 – 15:30 Spotlight Talks (7min + 2min Q&A per paper)

• Off-Policy Learning for Diversity-aware Candidate Retrieval in Two-stage Decisions
  Haruka Kiyohara, Rayhan Khanna and Thorsten Joachims.

• Sample-Free Almost-Exact Estimation of Plackett-Luce Propensities for Off-Policy Ranking Estimators
  Norman Knyazev and Harrie Oosterhuis.

• Unidentified and Confounded? Understanding Two-Tower Models for Unbiased Learning to Rank
  Philipp Hager, Onno Zoeter and Maarten de Rijke.

• Confounding is a Pervasive Problem in Real World Recommender Systems
  Alexander Merkov, David Rohde, Alexandre Gilotte and Benjamin Heymann.

• Offline Contextual Bandit with Counterfactual Sample Identification
  Alexandre Gilotte, Otmane Sakhi, Imad Aouali and Benjamin Heymann.

• RecGaze: The First Eye Tracking and User Interaction Dataset for Carousel Interfaces
  Santiago de Leon-Martinez, Jingwei Kang, Robert Moro, Maarten de Rijke, Branislav Kveton, Harrie Oosterhuis and Maria Bielikova.

15:30 – 16:00 Coffee Break 

16:00 – 16:30 Booster Session (2min per paper)

• Algorithm Adaptation Bias in Recommendation System Online Experiments
  Chen Zheng and Zhenyu Zhao.

• Optimal signals assignment for eBay VI page
  Matan Mandelbrod and Guy Feigenblat.

• Direct Profit Estimation Using Uplift Modeling under Cluster Network Interference
  Bram van den Akker.

• Investigating Action Embeddings for More Efficient Off-Policy Evaluation
  Maxime Wabartha, Kevin H. Wilson, R. David Evans, Hossein Sharifi-Noghabi and Tristan Sylvain.

• Off-Policy Learning in Large Action Spaces: Optimization Matters More Than Estimation
  Imad Aouali and Otmane Sakhi.

• Non-Linear Counterfactual Aggregate Optimization
  Benjamin Heymann and Otmane Sakhi.

• Adaptive Preference Aggregation for Recommender Systems
  Benjamin Heymann.

• Counterfactual Risk Minimization with IPS-Weighted BPR and Self-Normalized Evaluation in Recommender Systems
  Rahul Raja and Arpita Vats.

• Calibrated Recommendations with Contextual Bandits
  Diego Feijer, Himan Abdollahpouri, Sanket Gupta, Alexander Clare, Yuxiao Wen, Maria Dimakopoulou, Zahra Nazari, Kyle Kretschman, Mounia Lalmas and Todd Wasson.

• ACT: Automated Constraint Targeting for Multi-Objective Recommender Systems
  Daryl Chang, Yi Wu, Jennifer She, Li Wei and Lukasz Heldt.

• LLMs for estimating positional bias in logged interaction data
  Aleksandr Vladimirovich Petrov, Michael Murtagh and Karthik Nagesh.

• Epistemic Click Models Through Evidential Deep Learning
  Oscar Ramirez Milian and Harrie Oosterhuis.

• Adaptive Orchestration of Modular Generative Information Access Systems
  Mohanna Hoveyda, Harrie Oosterhuis, Arjen P. de Vries, Maarten de Rijke and Faegheh Hasibi.

16:30 – 17:30 Poster Session (Follow signs to poster area. CONSEQUENCES has poster spots #21-39.)

All times in local time zone of Prague, Czech Republic, (CEST, UTC+2).