Same Trajectory, Contradictory Rewards (ROBORMBENCH): Paraphrase Fragility in Vision Language Reward Models

Paper recorded by Signals 4 on 2026-09-04 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-04 on arXiv · recorded by Signals 4 on 2026-09-07

Category: cs.CL · 自然语言处理 · first seen 2026-09-07

Abstract

Vision-language models are increasingly used as reward functions for robotic learning, but this role requires paraphrase invariance: the same trajectory should receive the same reward under semantically equivalent goal descriptions. We show that current VLM reward models often violate this property. Paraphrasing the instruction alone can substantially change predicted progress scores, and can even

Read on arXiv →