Seeing Is Not Deciding: Can Multimodal LLMs Act as Effective CEOs?

• URL: http://arxiv.org/abs/2608.05864v1
• Date: Thu, 06 Aug 2026 10:43:33 GMT
• Title: Seeing Is Not Deciding: Can Multimodal LLMs Act as Effective CEOs?
• Abstract summary: We introduce C-SUITEBENCH, a controlled multimodal benchmark for executive business decisions.<n>We place nine models in the role of a chief executive officer and evaluate their decision-making ability.<n>We uncover a multimodal integration paradox: adding visual business information degrades constrained resource allocation for all nine models.
• Score: 53.05042283845008
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Large language models are increasingly applied as autonomous decision-making agents. However, in executive business decisions, existing benchmarks are limited to textonly settings. This makes it unclear whether models can perceive visual business evidence and effectively integrate it to improve decision quality. We introduce C-SUITEBENCH, a controlled multimodal benchmark that includes five decision tasks under paired text-only and multimodal conditions across 50 scenarios. We place nine frontier models in the role of a chief executive officer and evaluate their decision-making ability. Multimodal inputs consistently improve evidence-centric reasoning, with the largest and most reliable gains appearing in risk forecasting and board-facing justification. However, we uncover a multimodal integration paradox: adding visual business information degrades constrained resource allocation for all nine models, even as visual grounding itself improves. Ablation experiments reveal that this failure emerges from signal crowding, although each visual channel helps individually, their combination disrupts constraint satisfaction during decoding. These findings demonstrate that visual perception and constrained action are separable bottlenecks in multimodal agents, and that indiscriminate visual augmentation can harm high-stakes decision making, motivating selective grounding strategies for future executive AI systems.

Related papers

• Before Thinking, Learn to Decide: Proactive Routing for Efficient Visual Reasoning [87.18590662029432]
  We propose PRP, a Proactive Routing Paradigm that enables early decision-making by jointly evaluating the competence of both the draft and target models.<n>Our draft Rating Learning (DRL) equips the draft model with an internal confidence estimator, while Joint Rating Learning (JRL) predicts how well the target model can handle a given query.<n>These ratings enable fine-grained, instance-level textbfProactive Routing and substantially accelerate inference without compromising overall performance.
  arXiv  Detail & Related papers  (2026-06-29T12:30:24Z)
• Advancing Creative Physical Intelligence in Large Multimodal Models [62.56522271769017]
  We introduce MM-CreativityBench, a benchmark for affordance-grounded creative tool use in visually rich, physically constrained environments.<n>Our experiments show that current LMMs often fall short, not due to lack of generative capability, but because they do not sustain grounded exploration.<n>Motivated by this failure mode, we propose affordance-grounded alignment, which casts creative tool use as a preference learning problem.
  arXiv  Detail & Related papers  (2026-05-25T23:59:02Z)
• Decision-Making with Lightweight Confidence-Aware Language Model for Autonomous Driving [7.879160320881927]
  Large Language Models (LLMs) and Multimodal LLMs (MLLMs) have demonstrated immense potential in autonomous driving.<n>We propose a novel decision-making framework utilizing a lightweight confidence-aware language model.<n>We show that our approach achieves state-of-the-art (SOTA) success rates in both regular and long-tail scenarios.
  arXiv  Detail & Related papers  (2026-05-25T03:38:12Z)
• Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization [89.68681087743876]
  Multimodal Large Language Models (MLLMs) have incentivized models to think with images'' by actively invoking visual tools during multi-turn reasoning.<n>The common Reinforcement Learning practice of relying on outcome-based rewards ignores the fact that textual plausibility often masks executive failure.<n>We introduce Multimodal Agentic Policy Optimization (MAPO), bridging the gap between textual reasoning and visual actions generated by models within their Multimodal Chain-of-Thought.
  arXiv  Detail & Related papers  (2026-04-08T07:48:07Z)
• LLMs for Explainable Business Decision-Making: A Reinforcement Learning Fine-Tuning Approach [9.942833203981069]
  Large language models (LLMs) present an opportunity to generate natural-language explanations.<n>We introduce LEXMA, a reinforcement-learning-based fine-tuning framework that produces audience-appropriate explanations.<n>LEXMA yields significant improvements in predictive performance compared with other LLM baselines.
  arXiv  Detail & Related papers  (2025-12-10T04:16:31Z)
• Debate with Images: Detecting Deceptive Behaviors in Multimodal Large Language Models [25.61834023007555]
  We introduce MM-DeceptionBench, the first benchmark explicitly designed to evaluate multimodal deception.<n> MM-DeceptionBench characterizes how models strategically manipulate and mislead through combined visual and textual modalities.<n>We propose debate with images, a novel multi-agent debate monitor framework.
  arXiv  Detail & Related papers  (2025-11-29T06:39:36Z)
• DixitWorld: Evaluating Multimodal Abductive Reasoning in Vision-Language Models with Multi-Agent Dixit Gameplay [50.31585196187091]
  We introduce DixitWorld, a comprehensive evaluation suite designed to deconstruct multimodal abductive reasoning.<n>DIXITWORLD features two core components: DixitArena, a dynamic, multi-agent environment that evaluates hypothesis generation and hypothesis selection.<n>Results from DixitArena reveal distinct, role-dependent behaviors.
  arXiv  Detail & Related papers  (2025-10-11T08:48:48Z)
• Cantor: Inspiring Multimodal Chain-of-Thought of MLLM [83.6663322930814]
  We argue that converging visual context acquisition and logical reasoning is pivotal for tackling visual reasoning tasks. We propose an innovative multimodal CoT framework, termed Cantor, characterized by a perception-decision architecture. Our experiments demonstrate the efficacy of the proposed framework, showing significant improvements in multimodal CoT performance.
  arXiv  Detail & Related papers  (2024-04-24T17:59:48Z)
• NPHardEval4V: Dynamic Evaluation of Large Vision-Language Models with Effects of Vision [64.83085920775316]
  We introduce NPHardEval4V, a multimodal benchmark suite grounded in four classical NP-hard problems.<n>Each task is presented through a combination of structured visual layouts and textual prompts, designed to assess the ability of LVLMs to perform reasoning under visual-linguistic constraints.<n>Our results show that while these models perform reasonably well on perception-based inputs, they struggle with global optimization, abstraction, and constraint satisfaction.
  arXiv  Detail & Related papers  (2024-03-04T07:10:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.