← Back to all news
arXiv Machine Learning By Valentin Six, Frederik Panse, Mathis Fajeau, Lancelot Da Costa, Mridul Sharma, Alfonso Amayuelas, Tim Z. Xiao, David Hyland, Philipp Hennig, Bernhard Sch\"olkopf

Learning POMDP World Models from Observations with Language-Model Priors

Read the original on arXiv Machine Learning →

Learning POMDP World Models from Observations with Language-Model Priors proposes Pinductor, a method that uses a large language model to generate and refine candidate POMDP models from limited observation-action trajectories. Pinductor achieves performance comparable to privileged LLM-based methods while being up to 330 times more sample‑efficient than standard baselines like DreamerV3, and its effectiveness scales with LLM capability. The approach demonstrates that language‑model priors can substantially reduce the interaction cost required for learning internal world models under partial observability.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv AI
Sep 11

Belief-State Engine: Augmenting LLMs for Principled Planning Under Partial Observability

The paper introduces the Belief-State Engine (BSE), an inference module that supplies a large language model (LLM) with a Bayesian posterior over hidden states in a partially observable Markov decision process (POMDP). By keeping the raw action‑observation log hidden from the LLM, the BSE ensures the agent behaves as a sound Markov policy on the belief MDP, thereby inheriting classical POMDP optimality guarantees. Experiments on the Tiger POMDP and a red‑team attack‑graph task show that BSE‑augmented agents outperform six baselines in task return, belief calibration, and decision consistency.

By Arnab Chattopadhayay, Debdipta Halder
llmsagentsreinforcement-learning
More like this →
arXiv AI
Jun 16

PO-PDDL: Learning Symbolic POMDPs from Visual Demonstrations for Robot Planning Under Uncertainty

arXiv:2606. 15654v1 Announce Type: cross Abstract: Real-world robot task planning must operate under both stochastic action execution and partial observability, yet constructing Partially Observable Markov Decision Process (POMDP) models for real robotics domains remains difficult and labor-intensive.

By Wenjing Tang, Xuanjin Jin, Yuan Liu, Renming Huang, Cewu Lu, Panpan Cai
llmsreinforcement-learningrobotics
More like this →