Pith. sign in

REVIEW 4 cited by

Code as Reward: Empowering Reinforcement Learning with VLMs

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.04764 v1 pith:I74ZOJGM submitted 2024-02-07 cs.LG

David Venuto , Sami Nur Islam , Martin Klissarov , Doina Precup , Sherry Yang , Ankit Anand This is my paper
classification cs.LG
keywords vlmsrewardcodelearningrewardstrainingdensefeedback
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Pre-trained Vision-Language Models (VLMs) are able to understand visual concepts, describe and decompose complex tasks into sub-tasks, and provide feedback on task completion. In this paper, we aim to leverage these capabilities to support the training of reinforcement learning (RL) agents. In principle, VLMs are well suited for this purpose, as they can naturally analyze image-based observations and provide feedback (reward) on learning progress. However, inference in VLMs is computationally expensive, so querying them frequently to compute rewards would significantly slowdown the training of an RL agent. To address this challenge, we propose a framework named Code as Reward (VLM-CaR). VLM-CaR produces dense reward functions from VLMs through code generation, thereby significantly reducing the computational burden of querying the VLM directly. We show that the dense rewards generated through our approach are very accurate across a diverse set of discrete and continuous environments, and can be more effective in training RL policies than the original sparse environment rewards.

Discussion (0). Sign in to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning

    cs.RO 2026-03 conditional novelty 6.5 of 10

    A video-language model with per-timestep spatiotemporal CoT and dense progress prediction can serve as the sole reward for zero-shot online robot RL on 24 unseen manipulation tasks.

  2. Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills

    cs.RO 2026-08 conditional novelty 6.0 of 10

    A taxonomy of robot learning on a weights-versus-skills axis, with a five-rung self-improvement ladder whose top cell (feedback plus memory plus search) holds only a few recent systems.

  3. Reinforcement Learning for Machine Learning Engineering Agents

    cs.LG 2025-09 conditional novelty 6.0 of 10

    RL-trained Qwen2.5-3B outperforms prompted Claude-3.5-Sonnet and GPT-4o on 12 MLEBench tasks by an average of 22% and 24%, using two targeted RL modifications.

  4. ROAD: Responsibility-Oriented Reward Design for Reinforcement Learning in Autonomous Driving

    cs.LG 2025-05 reject novelty 6.0 of 10

    A responsibility-aware crash penalty, built from a traffic-law knowledge graph and a vision-language blame classifier, improves MetaDrive success rates and shifts reported collision blame away from the ego vehicle.

Pith tools