RM-VLM: From Natural-Language Task Descriptions to Vision-Grounded Reward Machines
cs.AI, cs.LG
Submitted: 2026-11-05
Updated: 2026-11-05
Terminology
Sources
- Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions
- FORM: Learning Expressive and Transferable First-Order Logic Reward Machines
- Vision-Language Models as a Source of Rewards
- End-to-End Abstraction-Based Control with LLM-Enhanced NL-to-LTL Translation
- ARM-FM: Automated Reward Machines via Foundation Models for Compositional Reinforcement Learning
- PlatoLTL: Scaling LTL-Guided Multi-Task RL
- Vision-Language Models as Success Detectors
- Gemma 4 Technical Report
- VICtoR: Learning Hierarchical Vision-Instruction Correlation Rewards for Long-horizon Manipulation
- Offline Reinforcement Learning with Implicit Q-Learning
- RoboReward: General-Purpose Vision-Language Reward Models for Robotics
- Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
- Ground-Compose-Reinforce: Grounding Language in Agentic Behaviours using Limited Data
- Reward Machines for Cooperative Multi-Agent Reinforcement Learning
- Adversarial Attacks to Reward Machine-based Reinforcement Learning
- Grounding LTL Tasks in Sub-Symbolic RL Environments for Zero-Shot Generalization
- Learning Robust Reward Machines from Noisy Labels
- OGBench: Benchmarking Offline Goal-Conditioned RL
- Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning
- Proximal Policy Optimization Algorithms
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents