Paper Detail
Hohin Kwan, Hongyu Li, Ray Zhang, Manyuan Zhang, Xianghao Kong, Anyi Rao, Jiahao Xie, Si Liu
Recent interest in multimodal large language models (MLLMs) raises a central question: can they reason over dynamic visual evidence rather than merely recognize objects or events in individual frames? This ability, which we refer to as video temporal-logical reasoning, requires models to maintain, update, and compose evidence as visual states evolve across frames. Existing video benchmarks often conflate this capability with scene complexity, static recognition, or uncontrolled temporal variation. To isolate this capability, we introduce Video-MME-Logical, a controlled benchmark organized around five temporal-logical operations: state tracking, sequential counting, temporal ordering, dynamic spatiality, and structural composition. The benchmark contains 25 fine-grained task categories generated with controlled object states, transitions, temporal dependencies, and logical compositions. It enables difficulty-controlled final-answer evaluation by varying temporal horizon and reasoning complexity, and supports intermediate-state diagnostics by verifying whether models recover the required logical reasoning trace before producing the final answer. Experiments with state-of-the-art MLLMs reveal a substantial human-model gap, especially as temporal-logical complexity increases. Supervised fine-tuning on up to 500K generated samples improves performance but remains insufficient to close the reasoning gap, positioning Video-MME-Logical as a scalable testbed for analyzing and improving temporal-logical reasoning in MLLMs.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{kwan2026video,
title = {Video-MME-Logical: A Controlled Diagnostic Benchmark for Video Temporal-Logical Reasoning},
author = {Hohin Kwan and Hongyu Li and Ray Zhang and Manyuan Zhang and Xianghao Kong and Anyi Rao and Jiahao Xie and Si Liu},
year = {2026},
abstract = {Recent interest in multimodal large language models (MLLMs) raises a central question: can they reason over dynamic visual evidence rather than merely recognize objects or events in individual frames? This ability, which we refer to as video temporal-logical reasoning, requires models to maintain, update, and compose evidence as visual states evolve across frames. Existing video benchmarks often conflate this capability with scene complexity, static recognition, or uncontrolled temporal variatio},
url = {https://huggingface.co/papers/2606.27828},
keywords = {multimodal large language models, video temporal-logical reasoning, temporal-logical operations, state tracking, sequential counting, temporal ordering, dynamic spatiality, structural composition, Video-MME-Logical, supervised fine-tuning, code available, huggingface daily},
eprint = {2606.27828},
archiveprefix = {arXiv},
}
{}