Theory, Practice, and Security of Foundation Models (Boninchi)

Contribution List

1. Welcome & intro

Slava Voloshynovskiy (UniGE)

29/09/2025, 09:00

2. Large Models in Low Precision: Power, Limits and Scaling Laws

Dan Alistarh ( ISTA, Austria)

29/09/2025, 09:30

Compression, searching and information-theoretic insights

4. Short-term attention enables long term memorization & Stochastic activations

Hervé Jégou (META)

29/09/2025, 11:00

Compression, searching and information-theoretic insights

5. AI through the Information-Theoretic lens

Slava Voloshynovskiy (UniGE)

29/09/2025, 13:10

Compression, searching and information-theoretic insights

8. TwinTURBO: Semi-Supervised Fine-Tuning of Foundation Models via Mutual Information Decompositions

Quetant Guillaume (UNIGE)

29/09/2025, 14:30

Compression, searching and information-theoretic insights

26. Llama4 in 4 bits

Brian Pulfer (University of Geneva)

29/09/2025, 15:00

46. Diffusion LM and Compression

Pavlo Molchanov (NVIDIA, USA)

29/09/2025, 15:50

37. The Watermark Awakens

Pierre Fernandez (Meta, France)

30/09/2025, 09:30

Security

47. The power of small language models

Pavlo Molchanov (NVIDIA, USA)

30/09/2025, 11:00

10. Handcrafted vs. Learned: A Comparative Evaluation of Digital Watermarking Techniques

Vitaliy Kinakh

30/09/2025, 13:40

Security

27. Attacking and Defending Vision Foundation Models

Yury Belousov

30/09/2025, 14:10

Security

12. TBD

Brian Pulfer (University of Geneva)

30/09/2025, 15:00

Security

11. TBD

Seyed Mohsen Moosavi-Dezfooli (Apple, CH)

30/09/2025, 15:50

Security

17. Round table - industrial perspective of model security and privacy

30/09/2025, 17:20

39. Boat trip to Yvoire

01/10/2025, 10:45

https://maps.app.goo.gl/5xWLTmPKiUcsABdm6

36. Boat trip -Return to Geneva

01/10/2025, 16:45

31. Generative World Models for Sample-Efficient Reinforcement Learning

Eloi Alonso, Vincent Micheli

02/10/2025, 09:30

GenAI and World Model

29. Energy-Based Transformers and the Future of Scaling

Alexi Gladstone (University of Illinois Urbana-Champaign/META, the USA)

02/10/2025, 11:00

GenAI and World Model

30. Mini-Bang Detection and Interpretation: Benchmarking Reasoning on Emergent and Open-Ended Dynamics

Andrey Ustuzanin (Constructor University, Germany)

02/10/2025, 13:40

GenAI and World Model

25. Improving mid-resource language performance in open LLMs

Hanna Yukhymenko (Hugging Face, CH)

02/10/2025, 14:50

GenAI and World Model

49. TBD

Yury Belousov

02/10/2025, 16:00

48. Binary Diffusion Probabilistic Model

Vitaliy Kinakh

02/10/2025, 16:30

32. Discussion & Brainstorm & Free time

02/10/2025, 17:00

34. DINOv3

Federico Baldassarre (Meta, France)

03/10/2025, 09:30

Vision Foundation Models \& Reasoning

45. How perception got solved (albeit unsatisfyingly)

Lucas Beyer, Xiaohua Zhaị

03/10/2025, 10:30

35. Reasoning by thinking compositionally

Emmanuel Abbe (EPFL / Apple, CH)

03/10/2025, 11:45

Vision Foundation Models \& Reasoning

42. World Models for Code

Taco Cohen (Meta, the UK)

03/10/2025, 14:00

Vision Foundation Models \& Reasoning

40. The free transformer

François Fleuret (University of Geneva /Meta, CH/France)

03/10/2025, 15:20

Vision Foundation Models \& Reasoning

41. Vision-language models: in-context learning and adaptation for radio-astronomy

Mariia Drozdova (Universite de Geneve (CH))

03/10/2025, 16:40

Vision Foundation Models \& Reasoning

43. Radio Astronomy in the Era of Large Models

Erica Lastufka (University of Geneva, CH)

03/10/2025, 17:10

Vision Foundation Models \& Reasoning

44. Summary/Conclusion/Outlook

Slava Voloshynovskiy (UniGE)

03/10/2025, 17:40

15. Discuss & Brainstorm & Free time

28. TBD

Pavlo Molchanov (NVIDIA, USA)

GenAI and World Model