8 papers
Total Variation Distance Estimation in Autoregressive Models
Eric Price, Kevin Tian, Zhiyang Xun +1
Modern LLM deployments use a number of implementation choices and inference optimizations (e.g., batching, custom kernels, and quantization) on top of fixed weights, so two engines…
Query Lower Bounds for Diffusion Sampling
Zhiyang Xun, Eric Price
Diffusion models generate samples by iteratively querying learned score estimates. A rapidly growing literature focuses on accelerating sampling by minimizing the number of score e…
Posterior Sampling by Combining Diffusion Models with Annealed Langevin Dynamics
Zhiyang Xun, Shivam Gupta, Eric Price
Given a noisy linear measurement of a distribution , and a good approximation to the prior , when can we sample from the posterior ? Posterior…
Avoiding Obfuscation with Prover-Estimator Debate
Jonah Brown-Cohen, Geoffrey Irving, Georgios Piliouras +3
Training powerful AI systems to exhibit desired behaviors hinges on the ability to provide accurate human supervision on increasingly complex tasks. A promising approach to this pr…
Spectral Guarantees for Adversarial Streaming PCA
Eric Price, Zhiyang Xun
In streaming PCA, we see a stream of vectors and want to estimate the top eigenvector of their covariance matrix. This is easier if the spectral…
Diffusion Posterior Sampling is Computationally Intractable
Shivam Gupta, Ajil Jalal, Aditya Parulekar +2
Diffusion models are a remarkably effective way of learning and sampling from a distribution . In posterior sampling, one is also given a measurement model and…