MINT combines symbolic trees with neural uncertainty estimation and LLM query curation to achieve near-expert planning performance by asking a small number of targeted questions that close knowledge gaps.
Learning from random demonstrations: Offline reinforcement learning with importance-sampled diffusion models.arXiv preprint arXiv:2405.19878
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.AI 2years
2026 2representative citing papers
A 13M-parameter intent-conditioned reward model trained on 398K multi-OS GUI steps scores candidate actions and lifts Agent S3 OSWorld success by 6.9 points without extra LLM calls.
citing papers explorer
-
MINT: Minimal Information Neuro-Symbolic Tree for Objective-Driven Knowledge-Gap Reasoning and Active Elicitation
MINT combines symbolic trees with neural uncertainty estimation and LLM query curation to achieve near-expert planning performance by asking a small number of targeted questions that close knowledge gaps.
-
IntentScore: Intent-Conditioned Action Evaluation for Computer-Use Agents
A 13M-parameter intent-conditioned reward model trained on 398K multi-OS GUI steps scores candidate actions and lifts Agent S3 OSWorld success by 6.9 points without extra LLM calls.