1W-MINDS, April 9: Flavio du Pin Calmon (Harvard University), Inference-Time Information Theory
Inference-Time Information Theory In this talk, we argue that "inference-time" Large Language Model (LLM) operation, where we interact with these models post-training without modifying their weights, is fertile ground for information-theoretic methods. We focus on one challenge in particular: watermarking LLM-generated text. Watermarks enable authentication of text provenance and help curb misuse of machine-generated content. We present recent results establishing a close connection between LLM watermarking and coding theory, showing that classical tools such as the Plotkin bound yield fundamental limits on watermark performance. This perspective also informs the design of two practical watermarks: SimplexWater and HeavyWater. We show that these watermarks achieve high detection accuracy with minimal impact on text quality, even in low-entropy tasks such as code generation. We also briefly survey other inference-time challenges that can be addressed with information theory, such as inference-time alignment. These results illustrate a broader opportunity: as LLMs increasingly serve as black-box components of more complex systems, information and coding theory offer a principled toolkit for shaping, verifying, and controlling their outputs.
![Yann LeCun's $1B Bet Against LLMs [Part 1]](https://i.ytimg.com/vi/kYkIdXwW2AE/hqdefault.jpg?sqp=-oaymwEjCNACELwBSFryq4qpAxUIARUAAAAAGAElAADIQj0AgKJDeAE=&rs=AOn4CLDbV4izF3i-wxevCVIn7FJjoy1vlA)
Yann LeCun's $1B Bet Against LLMs [Part 1]

Yann LeCun: World Models: Enabling the next AI revolution

1W-MINDS, March 26: Noam Razin (Princeton University), Understanding and Overcoming Pitfalls in...

Moody Gardens Penguin Cam LIVE | Penguin Habitat Stream at the Aquarium in Galveston, Texas

Training Sand to Think: Artificial General Intelligence & Future of Physics

I Was Right About AI

We're 99.9% sure this pattern is true, but no one can prove it

Creator of C++: Bell Labs, Negative Overhead Abstraction, Mistakes | Bjarne Stroustrup

The problem with pretending quantum mechanics makes sense | Sean Carroll

The Strange Math That Predicts (Almost) Anything

How ASML Makes Chips Faster With Its New $400 Million High NA Machine

Andrej Karpathy: From Vibe Coding to Agentic Engineering w/ Stephanie Zhan

Yann LeCun | Self-Supervised Learning, JEPA, World Models, and the future of AI

Quantum Computing for Computational Advantage

What rebuilding AlphaGo teaches us about self-play, RL, and future of LLMs - Eric Jang

Teenager Disproves 4 Decades Old Belief in Computing

Silicon Photonics and the Future of AI Scaling | John Bowers

1W-MINDS, Nov 6: Bohan Chen (Caltech) Learning Enhanced Ensemble Filters

Inference, Diffusion, World Models, and More | YC Paper Club

