12 papers
On the Generation Order in GFlowNets
Timofei Gritsaev, Evgeny Katasonov, Ildus Sadrtdinov +2
Sampling from discrete distributions specified by unnormalised probability functions is a classical statistical problem. GFlowNets address this problem by constructing objects sequ…
Amortised adaptive importance sampling
Arran Carter, Víctor Elvira, Esmeralda S. Whitammer
Monte Carlo sampling algorithms aim to produce samples from a target distribution with a given unnormalised probability density function. The high cost of running Monte Carlo algor…
Amortised Inference through One-Step Implicit Sampling
Vincent Pauline, Kirill Tamogashev, Arran Carter +3
Efficient sampling from Bayesian posterior distributions in high dimensions remains a hard problem, despite advances in theory and algorithms. Multiple steps are usually required t…
One for All, All for One: Coordinated Multi-Agent Diffusion Steering via Stochastic Optimal Control
Riccardo Barbano, Vincent Pauline, Runchang Li +6
Deep generative models often produce structured outputs composed of interacting components. Modelling these outputs with a single model requires learning both the component distrib…
KV-streams for Efficient Compaction in Agentic Reinforcement Learning
Emiliano Penaloza, Dane Malenfant, Dheeraj Vattikonda +15
Scaling the horizon of agentic LLMs is bottlenecked by the need to fit ever longer context traces in GPU memory. Context compaction has been the most popular mechanism to alleviate…
Strategically Diverse Sampling for Self-Training
Alexander Gurung, Esmeralda S. Whitammer, Mirella Lapata
Many LLM training and inference methods, including RL and test-time scaling, depend on repeated sampling, but benefit only when the responses meaningfully differ. Self-training fac…
Bayesian Symbolic Regression with Entropic Reinforcement Learning
Oussama Boussif, Mohammed Mahfoud, Younesse Kaddar +6
Symbolic regression is the problem of finding an algebraic expression describing a stochastic dependence of a target variable on a set of inputs. Unlike forms of regression that fi…
Path-dependent Discrete Amortized Inference
Tiago da Silva, Esmeralda S. Whitammer, Salem Lahlou
We consider the problem of sampling compositional and discrete objects from a given unnormalized posterior distribution. Notably, recent studies have shown that this problem can be…
Stop the Sampler! Classifier-Based Adaptive Stopping for Sampling Kernels
Kirill Korolev, Nikita Morozov, Stepan Pavlenko +2
Sampling from complex, unnormalized probability densities is a fundamental challenge in Bayesian inference and probabilistic modeling. While Markov chain Monte Carlo (MCMC) methods…
Optimal Transport Q-Learning for Flow Policy Steering and Acceleration
Andreas Sochopoulos, Esmeralda S. Whitammer, Nikolaos Tsagkas +3
Diffusion and flow policies have recently demonstrated remarkable performance in robotic applications by accurately capturing multimodal robot trajectory distributions, especially…
Structured Inference with Large Language Gibbs
Sanghyeok Choi, Henry Gouk, Esmeralda S. Whitammer
The knowledge encoded in large language models (LLMs) can serve as a substrate for structured reasoning over variables describing a complex world, but accessing this knowledge in a…
Imperfect World Models are Exploitable
Logan Mondal Bhamidipaty, Esmeralda S. Whitammer, David Abel +2
We propose a novel definition of model exploitation in reinforcement learning. Informally, a world model is exploitable if it implies that one policy should be strictly preferred o…