TrustRoboReward: Preference-Ordered Isotonic Score Editing for Multi-Paradigm Robot Reward Models
Yidong Wang, Yan Zhan, Ziteng Feng, Zhenyu Cui, Ziyi Zhou, Renzhao Liang, Jiaxuan Zhu, Zilei Yang, Yiran Zhao, Zhongkuan Mao, Bo Jia, Hanchu Ni, Chenggang Xie, Biao Liu, Yi Zhang, Yong Dai, Xiaozhu Ju, Wei Ye, Shikun Zhang
cs.AI
Submitted: 2026-08-09
Updated: 2026-08-11
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- RoboReward: General-Purpose Vision-Language Reward Models for Robotics
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- OpenHands: An Open Platform for AI Software Developers as Generalist Agents
- GPT-4 Technical Report
- Qwen3-VL Technical Report
- RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback
- Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
- OpenAI GPT-5 System Card
- PAI-Bench: A Comprehensive Benchmark For Physical AI
- Generative Reward Models
- CompassJudger-1: All-in-one Judge Model Helps Model Evaluation and Evolution
- Vision-Language Models as a Source of Rewards
- RewardBench: Evaluating Reward Models for Language Modeling
- Multimodal RewardBench: Holistic Evaluation of Reward Models for Vision Language Models
- Constitutional AI: Harmlessness from AI Feedback
- World Simulation with Video Foundation Models for Physical AI
Related papers
- 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
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems