arXiv cs.RO
Same Trajectory, Contradictory Rewards (ROBORMBENCH): Paraphrase Fragility in Vision Language Reward Models
같은 궤적, 상반된 보상 (ROBORMBENCH): 시각 언어 보상 모델에서의 패러프레이즈 취약성
No write-up yet — this entry is still just the auto-fetched hook. Aaron's own summary (200–500 words) will replace this note once he's written it.