ProactiveBench: Benchmarking Proactiveness in Multimodal Large Language Models

• URL: http://arxiv.org/abs/2603.19466v1
• Date: Thu, 19 Mar 2026 20:58:03 GMT
• Title: ProactiveBench: Benchmarking Proactiveness in Multimodal Large Language Models
• Abstract summary: We introduce ProactiveBench, a benchmark that tests proactiveness across different tasks.<n>We evaluate 22 MLLMs on ProactiveBench, showing that (i) they generally lack proactiveness; (ii) proactiveness does not correlate with model capacity; (iii) "hinting" at proactiveness yields only marginal gains.
• Score: 51.06179253173491
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Effective collaboration begins with knowing when to ask for help. For example, when trying to identify an occluded object, a human would ask someone to remove the obstruction. Can MLLMs exhibit a similar "proactive" behavior by requesting simple user interventions? To investigate this, we introduce ProactiveBench, a benchmark built from seven repurposed datasets that tests proactiveness across different tasks such as recognizing occluded objects, enhancing image quality, and interpreting coarse sketches. We evaluate 22 MLLMs on ProactiveBench, showing that (i) they generally lack proactiveness; (ii) proactiveness does not correlate with model capacity; (iii) "hinting" at proactiveness yields only marginal gains. Surprisingly, we found that conversation histories and in-context learning introduce negative biases, hindering performance. Finally, we explore a simple fine-tuning strategy based on reinforcement learning: its results suggest that proactiveness can be learned, even generalizing to unseen scenarios. We publicly release ProactiveBench as a first step toward building proactive multimodal models.

Related papers

• Activation-Based Active Learning for In-Context Learning: Challenges and Insights [4.090991964172345]
  In this paper, we test the hypothesis that model activations could provide a fine-grained signal to optimise the selection of in-context examples.<n>We present the most comprehensive analysis of activation-based deep active learning methods applied to in-context learning.
  arXiv  Detail & Related papers  (2026-06-03T17:39:29Z)
• IPIBench: Evaluating Interactive Proactive Intelligence of MLLMs under Continuous Streams [13.326802573561748]
  multimodal large language models (MLLMs) achieve strong performance on reactive question answering, but real-world streaming assistants require proactive reasoning over continuous visual inputs.<n>We introduce IPIBench, the first benchmark for evaluating Interactive Intelligence of MLLMs under streaming video settings.<n>We propose IPI-Agent, a training-free agentic framework with an interaction-control policy and a temporal-gating mechanism for stabilizing proactive triggering and coordinating multi-turn interactions.
  arXiv  Detail & Related papers  (2026-05-26T14:23:25Z)
• Anticipate and Learn: Unleashing Idle-Time Compute in Proactive Agents [51.46570338144688]
  ProAct is a proactive agent architecture that leverages idle-time compute to anticipate and fulfill likely upcoming user needs.<n>By analyzing evolving dialogue history together with persistent memory, ProAct predicts upcoming needs and iteratively acquires information.<n>MemBench evaluations confirm that ProAct achieves state-of-the-art reflective accuracy, underscoring its sustained and robust performance.
  arXiv  Detail & Related papers  (2026-05-25T15:47:21Z)
• From Reactive to Proactive: Assessing the Proactivity of Voice Agents via ProVoice-Bench [11.985605415045734]
  We introduce ProVoice-Bench, the first evaluation framework specifically designed for proactive voice agents.<n>By leveraging a multi-stage data synthesis pipeline, we curate 1,182 high-quality samples for rigorous testing.<n>Our evaluation of state-of-the-art Multimodal LLMs reveals a significant performance gap, particularly regarding over-triggering and reasoning capabilities.
  arXiv  Detail & Related papers  (2026-04-16T14:06:30Z)
• VisualActBench: Can VLMs See and Act like a Human? [47.16421650715271]
  Vision-Language Models (VLMs) have achieved impressive progress in perceiving and describing visual environments.<n>However, their ability to proactively reason and act based solely on visual inputs, without explicit textual prompts, remains underexplored.<n>We introduce a new task, Visual Action Reasoning, and propose VisualActBench, a large-scale benchmark comprising 1,074 videos and 3,733 human-annotated actions.
  arXiv  Detail & Related papers  (2025-12-10T18:36:18Z)
• When Seeing Is not Enough: Revealing the Limits of Active Reasoning in MLLMs [29.198301196459834]
  Multimodal large language models (MLLMs) have shown strong capabilities across a broad range of benchmarks.<n>Most existing evaluations focus on passive inference, where models perform step-by-step reasoning under complete information.<n>This raises a fundamental question: Can MLLMs actively acquire missing evidence under incomplete information?<n>We require the MLLMs to actively acquire missing evidence and iteratively refine decisions under incomplete information, by selecting a target image from a candidate pool without task-specific priors.<n>We evaluate 20 superior MLLMs and find that performance on active reasoning lags far behind it on passive settings, indicating substantial room for
  arXiv  Detail & Related papers  (2025-10-17T08:17:27Z)
• Active-O3: Empowering Multimodal Large Language Models with Active Perception via GRPO [63.140883026848286]
  Active vision refers to the process of actively selecting where and how to look in order to gather task-relevant information.<n>Recently, the use of Multimodal Large Language Models (MLLMs) as central planning and decision-making modules in robotic systems has gained extensive attention.
  arXiv  Detail & Related papers  (2025-05-27T17:29:31Z)
• Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling [51.38330727868982]
  We show how action chunking impacts the divergence between a learner and a demonstrator.<n>We propose Bidirectional Decoding (BID), a test-time inference algorithm that bridges action chunking with closed-loop adaptation.<n>Our method boosts the performance of two state-of-the-art generative policies across seven simulation benchmarks and two real-world tasks.
  arXiv  Detail & Related papers  (2024-08-30T15:39:34Z)
• Look Before You Decide: Prompting Active Deduction of MLLMs for Assumptive Reasoning [77.72128397088409]
  We show that most prevalent MLLMs can be easily fooled by the introduction of a presupposition into the question.<n>We also propose a novel reinforcement learning paradigm to encourage the model to actively perform composite deduction.
  arXiv  Detail & Related papers  (2024-04-19T15:53:27Z)
• Mind Your Outliers! Investigating the Negative Impact of Outliers on Active Learning for Visual Question Answering [71.15403434929915]
  We show that across 5 models and 4 datasets on the task of visual question answering, a wide variety of active learning approaches fail to outperform random selection. We identify the problem as collective outliers -- groups of examples that active learning methods prefer to acquire but models fail to learn. We show that active learning sample efficiency increases significantly as the number of collective outliers in the active learning pool decreases.
  arXiv  Detail & Related papers  (2021-07-06T00:52:11Z)

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.