Learning from Ranking Feedback: Improved Regret Bounds via Independence Preserving Rank Breaking

Nigel Strachan, Sattar Vakili, Matthijs Spaan, Julia Olkhovskaya

Paris Poster Session 6, Fri, Dec 11, 2:30 PM–4:30 PM, Paris Poster Hall · Published 2026

Dynamic Regret in Online Convex Optimization with Indicator Switching Costs

A meta-learning framework with randomized lazy FTRL and movement-aware mixing achieves near-optimal dynamic regret with indicator switching costs and adapts to both switch counts and path length without prior knowledge.

Naram Mhaisen, George Iosifidis

Paris Poster Session 2, Wed, Dec 9, 5:00 PM–7:00 PM, Paris Poster Hall · Published 2026

CompleteRXN: Toward Completing Open Chemical Reaction Databases

CompleteRXN introduces a benchmark for completing incomplete chemical reaction databases, showing models reach high accuracy on benchmark splits but degrade substantially on uncurated real-world data.

Gabriel Vogel, Minouk Noordsij, Evgeny A Pidko, Jana M. Weber

Paris Poster Session 6, Fri, Dec 11, 2:30 PM–4:30 PM, Paris Poster Hall · Published 2026

Invaria: Learning Scale and Density Invariance in Point Clouds via Next-Resolution Prediction

Invaria learns scale and density invariant point cloud features via next-resolution prediction, boosting low-resolution ScanNet mIoU by 56% with a smaller model and fewer tokens.

Chun-Peng Chang, Shaoxiang Wang, Alain Pagani, Dariu Gavrila and 1 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Tight Generalization Bounds for Noiseless Inverse Optimization

Noiseless inverse optimization achieves tight O(d/T) generalization and regret bounds, with parameter-free algorithms matching adversarial lower bounds.

Sayedpouria Fatemi, Hoomaan Maskan, Suvrit Sra, Peyman Mohajerin Esfahani

Paris Poster Session 1, Wed, Dec 9, 12:30 PM–2:30 PM, Paris Poster Hall · Published 2026