VideoReasonBench: Can MLLMs Perform Vision-Centric Complex Video Reasoning?

• URL: http://arxiv.org/abs/2505.23359v1
• Date: Thu, 29 May 2025 11:33:43 GMT
• Title: VideoReasonBench: Can MLLMs Perform Vision-Centric Complex Video Reasoning?
• Authors: Yuanxin Liu, Kun Ouyang, Haoning Wu, Yi Liu, Lin Sui, Xinhao Li, Yan Zhong, Y. Charles, Xinyu Zhou, Xu Sun,
• Abstract summary: Long chain-of-thought (CoT) reasoning can significantly enhance the performance of large language models (LLMs) on complex tasks.<n>Recent efforts have proposed benchmarks aimed at video reasoning, but tasks are often knowledge-driven and do not rely heavily on visual content.<n>We introduce VideoReasonBench, a benchmark designed to evaluate vision-centric, complex video reasoning.
• Score: 18.9270920369958
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent studies have shown that long chain-of-thought (CoT) reasoning can significantly enhance the performance of large language models (LLMs) on complex tasks. However, this benefit is yet to be demonstrated in the domain of video understanding, since most existing benchmarks lack the reasoning depth required to demonstrate the advantages of extended CoT chains. While recent efforts have proposed benchmarks aimed at video reasoning, the tasks are often knowledge-driven and do not rely heavily on visual content. To bridge this gap, we introduce VideoReasonBench, a benchmark designed to evaluate vision-centric, complex video reasoning. To ensure visual richness and high reasoning complexity, each video in VideoReasonBench depicts a sequence of fine-grained operations on a latent state that is only visible in part of the video. The questions evaluate three escalating levels of video reasoning skills: recalling observed visual information, inferring the content of latent states, and predicting information beyond the video. Under such task setting, models have to precisely recall multiple operations in the video, and perform step-by-step reasoning to get correct final answers for these questions. Using VideoReasonBench, we comprehensively evaluate 18 state-of-the-art multimodal LLMs (MLLMs), finding that most perform poorly on complex video reasoning, e.g., GPT-4o achieves only 6.9% accuracy, while the thinking-enhanced Gemini-2.5-Pro significantly outperforms others with 56.0% accuracy. Our investigations on "test-time scaling" further reveal that extended thinking budget, while offering none or minimal benefits on existing video benchmarks, is essential for improving the performance on VideoReasonBench.

Related papers

• Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning [29.811030252357195]
  multimodal large language models (MLLMs) are crucial for downstream tasks like video question answering and temporal grounding.<n>We propose Video Intelligence via Tool-Augmented Learning (VITAL), a novel end-to-end agentic video reasoning framework.
  arXiv  Detail & Related papers  (2025-08-06T13:03:21Z)
• Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning? [56.06537213958482]
  We present Video-Holmes, a benchmark designed to evaluate the complex video reasoning capabilities of MLLMs.<n>Video-Holmes consists of 1,837 questions derived from 270 manually annotated suspense short films.<n>Our comprehensive evaluation of state-of-the-art MLLMs reveals that, while these models generally excel at visual perception, they encounter substantial difficulties with integrating information.
  arXiv  Detail & Related papers  (2025-05-27T16:05:01Z)
• SAMA: Towards Multi-Turn Referential Grounded Video Chat with Large Language Models [80.3895950009792]
  Achieving fine-grained-temporal understanding in videos remains a major challenge for current Video Large Multimodels (Video LMMs)<n>We contribute in three core aspects: dataset, model, and benchmark.<n>First, we introduce SAMA-239K, a large-scale dataset comprising 15K videos specifically to enable joint learning of video understanding, grounding, and multi-turn video chat.<n>Second, we propose the SAMA model, which incorporates a versatile-temporal context aggregator and a Segment Model to jointly enhance fine-grained video comprehension and precise grounding capabilities.
  arXiv  Detail & Related papers  (2025-05-24T18:13:16Z)
• video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model [33.70837005629285]
  We propose video-SALMONN-o1, the first open-source reasoning-enhanced audio-visual LLM designed for general video understanding tasks.<n>We develop a reasoning-intensive dataset featuring challenging audio-visual questions with step-by-step solutions.<n>We also introduce RivaBench, the first reasoning-intensive video understanding benchmark, featuring over 4,000 high-quality, expert-curated question-answer pairs.
  arXiv  Detail & Related papers  (2025-02-17T13:07:40Z)
• SCBench: A Sports Commentary Benchmark for Video LLMs [19.13963551534595]
  We develop a benchmark for sports video commentary generation for Video Large Language Models (Video LLMs)<n>$textbfSCBench$ is a six-dimensional metric specifically designed for our task, upon which we propose a GPT-based evaluation method.<n>Our results found InternVL-Chat-2 achieves the best performance with 5.44, surpassing the second-best by 1.04.
  arXiv  Detail & Related papers  (2024-12-23T15:13:56Z)
• STEP: Enhancing Video-LLMs' Compositional Reasoning by Spatio-Temporal Graph-guided Self-Training [87.58996020705258]
  Video Large Language Models (Video-LLMs) have recently shown strong derivation in basic video understanding tasks.<n>Video-LLMs struggle with compositional reasoning that requires multi-step explicit-temporal inference across object relations, interactions and events.<n>We propose STEP, a novel graph-guided self-training method that enables VideoLLMs to generate reasoning-rich finetuning data from any raw videos to improve itself.
  arXiv  Detail & Related papers  (2024-11-29T11:54:55Z)
• VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection [61.54044967253421]
  We introduce VideoEspresso, a novel dataset that features VideoQA pairs preserving essential spatial details and temporal coherence. Our construction pipeline employs a semantic-aware method to reduce redundancy, followed by generating QA pairs using GPT-4o. We propose a Hybrid LVLMs Collaboration framework, featuring a Frame Selector and a two-stage instruction fine-tuned reasoning LVLM.
  arXiv  Detail & Related papers  (2024-11-22T08:33:36Z)
• E.T. Bench: Towards Open-Ended Event-Level Video-Language Understanding [57.630136434038384]
  We introduce E.T. Bench (Event-Level & Time-Sensitive Video Understanding Benchmark), a large-scale benchmark for open-ended event-level video understanding. We extensively evaluated 8 Image-LLMs and 12 Video-LLMs on our benchmark, and the results reveal that state-of-the-art models for coarse-level (video-level) understanding struggle to solve our fine-grained tasks. Our simple but effective solution demonstrates superior performance in multiple scenarios.
  arXiv  Detail & Related papers  (2024-09-26T17:53:04Z)
• LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding [41.9477837230283]
  LongVideoBench is a question-answering benchmark that features video-language interleaved inputs up to an hour long. Our benchmark includes 3,763 varying-length web-collected videos with their subtitles across diverse themes. We formulate a novel video question-answering task termed referring reasoning.
  arXiv  Detail & Related papers  (2024-07-22T16:00:55Z)
• MMBench-Video: A Long-Form Multi-Shot Benchmark for Holistic Video Understanding [67.56182262082729]
  We introduce MMBench-Video, a quantitative benchmark to rigorously evaluate large vision-language models (LVLMs) in video understanding. MMBench-Video incorporates lengthy videos from YouTube and employs free-form questions, mirroring practical use cases. The benchmark is meticulously crafted to probe the models' temporal reasoning skills, with all questions human-annotated according to a carefully constructed ability taxonomy.
  arXiv  Detail & Related papers  (2024-06-20T17:26:01Z)
• VideoVista: A Versatile Benchmark for Video Understanding and Reasoning [46.838692817107116]
  We present VideoVista, a video QA benchmark that integrates challenges across diverse content categories, durations, and abilities. VideoVista comprises 25,000 questions derived from 3,400 videos spanning 14 categories (e.g., Howto, Film, and Entertainment) with durations ranging from a few seconds to over 10 minutes. It encompasses 19 types of understanding tasks (e.g., anomaly detection, interaction understanding) and 8 reasoning tasks (e.g., logical reasoning, causal reasoning)
  arXiv  Detail & Related papers  (2024-06-17T08:09:00Z)
• VELOCITI: Benchmarking Video-Language Compositional Reasoning with Strict Entailment [19.313541287648473]
  We introduce VELOCITI, a benchmark to study Video-LLMs by disentangling and assessing the comprehension of agents.<n>We adopt the Video-Language Entailment setup and propose StrictVLE that requires correct classification (rather than ranking) of the positive and negative caption.<n>Results show that action understanding lags behind agents, and negative captions created using entities appearing in the video perform worse than those obtained from pure text manipulation.
  arXiv  Detail & Related papers  (2024-06-16T10:42:21Z)

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.