5 papers
Selective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers
Giovanni Bonetta, Matteo Merler, Davide Zago +2
Vision-Language Models (VLMs) provide useful priors for interactive decision-making, but using them directly as policies is expensive and brittle: they must be queried at every ste…
QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents
Sergio Hernández-Gutiérrez, Matteo Merler, Ilze Amanda Auzina +3
LLM agents increasingly act over long horizons, where a single trajectory can contain hundreds or thousands of actions. In these settings, outcome-only rewards provide too sparse g…
DECSELFMASK: Leveraging Unlabeled Text via Self-Relevance-Guided Masking for Decoder-Only Classification
Pietro Ferrazzi, Matteo Merler, Giovanni Bonetta +2
Classification tasks require annotated data, which can often be expensive, time-consuming, or even unfeasible to collect. This is the case of the medical domain, where large datase…
ViPlan: A Benchmark for Visual Planning with Symbolic Predicates and Vision-Language Models
Matteo Merler, Nicola Dainese, Minttu Alakuijala +5
Integrating Large Language Models with symbolic planners is a promising direction for obtaining verifiable and grounded plans, with recent works extending this idea to visual domai…
Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search
Nicola Dainese, Matteo Merler, Minttu Alakuijala +1
In this work we consider Code World Models, world models generated by a Large Language Model (LLM) in the form of Python code for model-based Reinforcement Learning (RL). Calling c…