Same Trajectory, Contradictory Rewards (ROBORMBENCH): Paraphrase Fragility in Vision Language Reward Models
New ROBORMBENCH benchmark shows vision-language reward models can flip robot success/failure judgments when goal instructions are paraphrased.
The authors show that paraphrasing the instruction alone can substantially change progress scores from VLM reward models, even flipping identical robot trajectories between failure and success. ROBORMBENCH comprises 2,390 real-robot trajectories with ground-truth progress labels and 21,673 verified paraphrases covering lexical, syntactic, and action-goal rewrites. Instability is widespread across proprietary and open-source VLMs, grows with more divergent rewrites, and is not reliably reduced by scale or explicit reasoning, while trajectory-grounded dedicated reward models are markedly more stable.
- Paraphrasing instructions alone can flip identical robot behavior between failure and success in VLM reward models.
- Benchmark includes 2,390 real-robot trajectories, ground-truth labels, and 21,673 verified paraphrases.
- Instability worsens with more divergent rewrites; scale and explicit reasoning do not reliably help.
- Trajectory-grounded dedicated reward models are substantially more stable.
From the original article
1 min read at arxiv.orgVision-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.
By Wonje Jeung · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.