# Agent-RewardBench: Towards a Unified Benchmark for Reward Modeling across Perception, Planning, and Safety in Real-World Multimodal Agents

arXiv: https://arxiv.org/abs/2506.21252 · HTML: https://mlsanity.com/paper/2506.21252/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 32 of its 45 references that are in the map.
- First posted on arXiv 26 Jun 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)
- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)
- LLM alignment / RLHF / preference optimisation / [Process reward models for reasoning](https://mlsanity.com/alignment/process-reward-models-for-reasoning/)

## Builds on (most cited first, 30 of 32)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Mind2Web: Towards a Generalist Agent for the Web](https://mlsanity.com/paper/2306.06070/) (2023)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites](https://mlsanity.com/paper/2404.16821/) (2024)
- [SeeClick: Harnessing GUI Grounding for Advanced Visual GUI Agents](https://mlsanity.com/paper/2401.10935/) (2024)
- [AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents](https://mlsanity.com/paper/2405.14573/) (2024)
- [Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents](https://mlsanity.com/paper/2302.01560/) (2023)
- [TravelPlanner: A Benchmark for Real-World Planning with Language Agents](https://mlsanity.com/paper/2402.01622/) (2024)
- [AgentTuning: Enabling Generalized Agent Abilities for LLMs](https://mlsanity.com/paper/2310.12823/) (2023)
- [FireAct: Toward Language Agent Fine-tuning](https://mlsanity.com/paper/2310.05915/) (2023)
- [Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents](https://mlsanity.com/paper/2403.02502/) (2024)
- [OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task Synthesis](https://mlsanity.com/paper/2412.19723/) (2024)
- [Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents](https://mlsanity.com/paper/2411.06559/) (2024)
- [Tree Search for Language Model Agents](https://mlsanity.com/paper/2407.01476/) (2024)
- [WebPilot: A Versatile and Autonomous Multi-Agent System for Web Task Execution with Strategic Exploration](https://mlsanity.com/paper/2408.15978/) (2024)
- [Agent Lumos: Unified and Modular Training for Open-Source Language Agents](https://mlsanity.com/paper/2311.05657/) (2023)
- [Towards End-to-End Embodied Decision Making via Multi-modal Large Language Model: Explorations with GPT4-Vision and Beyond](https://mlsanity.com/paper/2310.02071/) (2023)
- [VSP: Assessing the dual challenges of perception and reasoning in spatial planning tasks for VLMs](https://mlsanity.com/paper/2407.01863/) (2024)
- [Large Language Models for Planning: A Comprehensive and Systematic Survey](https://mlsanity.com/paper/2505.19683/) (2025)
- [PCA-Bench: Evaluating Multimodal Large Language Models in Perception-Cognition-Action Chain](https://mlsanity.com/paper/2402.15527/) (2024)
- [Enhancing Decision-Making for LLM Agents via Step-Level Q-Value Models](https://mlsanity.com/paper/2409.09345/) (2024)
- [RAG-RewardBench: Benchmarking Reward Models in Retrieval Augmented Generation for Preference Alignment](https://mlsanity.com/paper/2412.13746/) (2024)
- [Unlocking the Future: Exploring Look-Ahead Planning Mechanistic Interpretability in Large Language Models](https://mlsanity.com/paper/2406.16033/) (2024)

## Built on by (most cited first, 20 of 20)

- [AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents](https://mlsanity.com/paper/2603.14465/) (2026)
- [CUARewardBench: A Benchmark for Evaluating Reward Models on Computer-using Agent](https://mlsanity.com/paper/2510.18596/) (2025)
- [Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification](https://mlsanity.com/paper/2507.11662/) (2025)
- [AJ-Bench: Benchmarking Agent-as-a-Judge for Environment-Aware Evaluation](https://mlsanity.com/paper/2604.18240/) (2026)
- [Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form Preferences](https://mlsanity.com/paper/2510.23451/) (2025)
- [IF-RewardBench: Benchmarking Judge Models for Instruction-Following Evaluation](https://mlsanity.com/paper/2603.04738/) (2026)
- [OS-Themis: A Scalable Critic Framework for Generalist GUI Rewards](https://mlsanity.com/paper/2603.19191/) (2026)
- [Aligning Agents via Planning: A Benchmark for Trajectory-Level Reward Modeling](https://mlsanity.com/paper/2604.08178/) (2026)
- [OS-Oracle: A Comprehensive Framework for Cross-Platform GUI Critic Models](https://mlsanity.com/paper/2512.16295/) (2025)
- [Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning](https://mlsanity.com/paper/2606.27330/) (2026)
- [Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning](https://mlsanity.com/paper/2608.03571/) (2026)
- [Locating Hidden Failures Makes Long-Horizon Agents More Reliable](https://mlsanity.com/paper/2609.17930/) (2026)
- [Empowering Real-World: A Survey on the Technology, Practice, and Evaluation of LLM-driven Industry Agents](https://mlsanity.com/paper/2510.17491/) (2025)
- [Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety](https://mlsanity.com/paper/2603.18245/) (2026)
- [SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs](https://mlsanity.com/paper/2608.03573/) (2026)
- [The Deliberative Illusion: Diagnosing Factual Attrition and Stance Homogenization in Multi-Agent LLM Deliberation](https://mlsanity.com/paper/2606.03032/) (2026)
- [Evaluating Reward Model Generalization via Pairwise Maximum Discrepancy Competitions](https://mlsanity.com/paper/2601.16987/) (2026)
- [ATLAS: Dual-Horizon Diagnostic Evaluation for Industrial Tool-Use Agents](https://mlsanity.com/paper/2608.30685/) (2026)
- [CRISP: Cultural Reward Modeling for Implicit Situated Propriety](https://mlsanity.com/paper/2609.34345/) (2026)
- [DUET: Co-Evolving Solver and Grader Agents](https://mlsanity.com/paper/2610.04087/) (2026)

## Abstract (arXiv)

As Multimodal Large Language Models (MLLMs) advance, multimodal agents show promise in real-world tasks like web navigation and embodied intelligence. However, due to limitations in a lack of external feedback, these agents struggle with self-correction and generalization. A promising approach is to use reward models as external feedback, but there is no clear on how to select reward models for agents. Thus, there is an urgent need to build a reward bench targeted at agents. To address these challenges, we propose Agent-RewardBench, a benchmark designed to evaluate reward modeling ability in MLLMs. The benchmark is characterized by three key features: (1) Multiple dimensions and real-world agent scenarios evaluation. It covers perception, planning, and safety with 7 scenarios; (2) Step-level reward evaluation. It allows for the assessment of agent capabilities at the individual steps of a task, providing a more granular view of performance during the planning process; and (3) Appropriately difficulty and high-quality. We carefully sample from 10 diverse models, difficulty control to maintain task challenges, and manual verification to ensure the integrity of the data. Experiments demonstrate that even state-of-the-art multimodal models show limited performance, highlighting the need for specialized training in agent reward modeling. Code is available at github.