Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.23698 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916974603272192 |
|---|---|
| author | Hu, Zhe Ren, Yixiao Liu, Guanzhong Li, Jing Yin, Yu |
| author_facet | Hu, Zhe Ren, Yixiao Liu, Guanzhong Li, Jing Yin, Yu |
| contents | Multimodal Large Language Models (MLLMs) show promising results for embodied agents in operating meaningfully in complex, human-centered environments. Yet, evaluating their capacity for nuanced, human-like reasoning and decision-making remains challenging. In this work, we introduce VIVA+, a cognitively grounded benchmark for evaluating the reasoning and decision-making of MLLMs in human-centered situations. VIVA+ consists of 1,317 real-world situations paired with 6,373 multiple-choice questions, targeting three core abilities for decision-making: (1) Foundational Situation Comprehension, (2) Context-Driven Action Justification, and (3) Reflective Reasoning. Together, these dimensions provide a systematic framework for assessing a model's ability to perceive, reason, and act in socially meaningful ways. We evaluate the latest commercial and open-source models on VIVA+, where we reveal distinct performance patterns and highlight significant challenges. We further explore targeted training and multi-step reasoning strategies, which yield consistent performance improvements. Finally, our in-depth analysis highlights current model limitations and provides actionable insights for advancing MLLMs toward more robust, context-aware, and socially adept decision-making in real-world settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_23698 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VIVA+: Human-Centered Situational Decision-Making Hu, Zhe Ren, Yixiao Liu, Guanzhong Li, Jing Yin, Yu Computation and Language Multimodal Large Language Models (MLLMs) show promising results for embodied agents in operating meaningfully in complex, human-centered environments. Yet, evaluating their capacity for nuanced, human-like reasoning and decision-making remains challenging. In this work, we introduce VIVA+, a cognitively grounded benchmark for evaluating the reasoning and decision-making of MLLMs in human-centered situations. VIVA+ consists of 1,317 real-world situations paired with 6,373 multiple-choice questions, targeting three core abilities for decision-making: (1) Foundational Situation Comprehension, (2) Context-Driven Action Justification, and (3) Reflective Reasoning. Together, these dimensions provide a systematic framework for assessing a model's ability to perceive, reason, and act in socially meaningful ways. We evaluate the latest commercial and open-source models on VIVA+, where we reveal distinct performance patterns and highlight significant challenges. We further explore targeted training and multi-step reasoning strategies, which yield consistent performance improvements. Finally, our in-depth analysis highlights current model limitations and provides actionable insights for advancing MLLMs toward more robust, context-aware, and socially adept decision-making in real-world settings. |
| title | VIVA+: Human-Centered Situational Decision-Making |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.23698 |