Archimedes at NeurIPS 2026

Archimedes members will present 23 papers at the Conference on Neural Information Processing Systems (NeurIPS 2026), one of the world’s leading AI conferences: one oral, three spotlights and 19 posters.

Oral

• MaxSketch: Robust Distinct Counting in Streams via Random Projections
  Nikos Tsikouras, Constantine Caramanis, Christos Tzamos.

Spotlights

1. fmxcoders: Factorized Masked Crosscoders for Cross-Layer Feature Discovery
   A. D. Demou, P. Koromilas, J. Oldfield, Yannis Panagakis, M. Nicolaou.
2. Online Fair Division Meets Reordering Buffers
   Georgios (Yorgos) Amanatidis, G. Giaconi, Vangelis Markakis, Nicos Protopapas.
3. The Long-Run Distribution of Regularized Learning in Non-Concave Games: A Large Deviations Approach
   W. Azizian, P. L. Cauvin, F. Lutzeler, J. Malick, Panayotis Mertikopoulos.

Posters

1. Efficient Computation and Best-Response Dynamics in Anonymous Two-Action Games with Linear Utilities
   M. Fasoulakis, Vangelis Markakis, I. Panageas, Christodoulos Santorinaios, J. Yan.
2. GATE-AD: Graph Attention Network Encoding for Few-Shot Industrial Visual Anomaly Detection
   A. Psiris, Yannis Panagakis, Maria Vakalopoulou, Georgios Th. Papadopoulos.
3. Interpretability-by-Design with Accurate Locally Additive Models and Conditional Feature Effects
   V. Gkolemis, L. Kavouras, D. Kyriakopoulos, K. Tsopelas, D. Rontogiannis, G. Casalicchio, T. Dalamagas, Christos Diou.
4. Learning-Augmented Coordination Mechanisms
   George Christodoulou, Vasilis Christoforidis, Alkmini Sgouritsa, Ioannis Vlachos.
5. NeuroRVQ: Multi-Scale Biosignal Tokenization for Generative Foundation Models
   K. Barmpas, N. Lee, D. Halatsis, W. Raftery, Yannis Panagakis, D. Adamos, N. Laskaris, A. Koliousis, D. Farina, S. Zafeiriou.
6. Online Learning on Hidden-Convex Losses via Algorithmic Equivalence: Optimal Regret, Geometric Barrier, and Bandit Feedback
   A. Barakat, Andreas Kontogiannis, Vasilis Pollatos, I. Panageas, Antonios Varvitsiotis.
7. Rare Events, Real Signals: Functional Ensembles as a Computational Substrate in Deep Networks
   A. Aravind, K. Ladakis, M. Alexios Savaglio, S. M. Smirnakis, Maria Papadopouli.
8. The Grounding Gap: How LLMs Anchor the Meaning of Abstract Concepts Differently from Humans
   Odysseas Chlapanis, O. M. Mastromichalakis, C. H. Papadimitriou.
9. What to Predict for Efficient Scheduling on Parallel Machines
   E. Bampis, B. Escoffier, Dimitris Fotakis, G. Mitropoulos, M. Xefteris.
10. Scaling Multi-Teacher Distillation for Digital Pathology
    S. Boutaj, P. Marza, V. Belagali, D. Samaras, Maria Vakalopoulou, S. Christodoulidis.
11. Class Adaptive Conformal Training
    B. E. Marani, J. S. Rodríguez, I. B. Ayed, Maria Vakalopoulou, S. Christodoulidis, J. Dolz.
12. A Unified Framework for Uniform-Price Resource Allocation Mechanisms
    I. Caragiannis, Dimitris Fotakis, S. Skoulakis.
13. UniGS: Unified Geometry-Aware Gaussian Splatting for Multimodal Rendering
    Y. Xie, Z. Huang, J. Jiao, Dimitrios Kanoulas, J. Ma.
14. State-Action Selection in General Stochastic Games under Regularized Policy Gradient Methods
    O. Abbadi, R. Laraki, Panayotis Mertikopoulos.
15. Adapting to Conflict: Equilibrium Structure and Adaptive Learning in Harmonic Games
    D. Legacci, Panayotis Mertikopoulos, B. Pradelski.
16. When and Why is Optimistic Multiplicative Weights Slow? The Geometry of Energy Dissipation
    J. Lazarsfeld, A. Barakat, G. Piliouras, Antonios Varvitsiotis, A. Wibisono.
17. Multi-agent online control with adversarial disturbances
    A. Barakat, J. Lazarsfeld, G. Piliouras, Antonios Varvitsiotis.
18. Capturing membrane dynamics and spike timing of human neurons at scale using neural operators
    L. Ghafourpour, C. Zhao, V. Duruisseaux, B. Tolooshams, P. H. Wong, Costas Anastassiou, A. Anandkumar.

Datasets and Benchmarks Track

19. FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs
    D. Restrepo, C. Wu, L. F. Nakayama, M. L. Martins, S. Christodoulidis, Maria Vakalopoulou, E. Ferrante.

Congratulations to all of our members and collaborators whose hard work made this possible.

Explore more publications by Archimedes members: archimedesai.gr/en/publications