VidOmni-Bench is a new benchmark for fine‑grained video understanding that asks models to verify whether each event in dense video captions is supported by the video. It contains 500 videos covering five complexity types and durations from 4 seconds to 90 minutes, and uses human‑verified sentence‑level labels to create hard negatives. Experiments show that Video‑LLMs often hallucinate events, struggle to detect incorrect descriptions, and exhibit varying weaknesses depending on video complexity and duration.
arXiv:2609.09973v1 Announce Type: new Abstract: Evaluating video captioning remains a critical challenge for Visual Large Language Models (VLLMs). Existing metrics primarily rely on matching generate...
The paper introduces CodeCap-Bench, a benchmark for evaluating whether executable code can serve as an effective video caption. It compares code captions to natural language captions on tasks such as question answering and video reconstruction, finding that code captions yield higher reasoning accuracy and reconstruction fidelity. The study also assesses current multimodal large language models, noting that while proprietary models show promise, they still struggle with complex temporal dynamics, and open‑source models lag significantly.
arXiv:2510. 14904v4 Announce Type: replace-cross Abstract: Dense Video Object Captioning (DVOC) is the task of jointly detecting, tracking, and captioning object trajectories in a video, requiring the ability to understand spatio-temporal details and describe them in natural language.
MEV is a multi‑event video dataset created to support long‑take video generation. It contains high‑quality real‑world videos where subjects perform distinct, non‑overlapping events with precise temporal boundaries and fine‑grained procedural annotations. The dataset includes a comprehensive annotation pipeline that segments videos into temporally grounded events and generates structured captions at both global and event levels, enabling event‑centric evaluation and improved long‑take event transitions in generation experiments.
The paper evaluates 14 state‑of‑the‑art text‑to‑video retrieval models on three major datasets using a unified framework. It finds that short, clear, and simple captions—especially those describing single actions or color attributes—yield higher recall, while complex, multi‑step events or fine‑grained scene descriptions remain difficult for all models. Attention‑driven architectures better handle temporally dependent or multi‑step queries, whereas dual‑encoder and multimodal fusion models excel mainly on simpler captions, and cross‑dataset generalization improves with larger, more diverse caption sets but generative captions do not consistently boost accuracy.
The paper introduces Seeing Before Synthesizing (SBS), a weakly-supervised dense video captioning framework that uses a vision‑language model to generate frame‑level narratives for gaps between events and detect transitions based on semantic changes. SBS refines temporal masks by aligning transition points with vision‑language cues, rather than relying on rigidly placed synthetic captions. Experiments on ActivityNet Captions and YouCook2 show that SBS achieves state‑of‑the‑art results in both captioning and localization tasks.
arXiv:2606. 24636v1 Announce Type: new Abstract: Cinematographic captioning aims to describe how a video is filmed using professional film-language concepts such as camera movement, shot size, depth of field, composition, and shooting angle.
arXiv:2604. 10024v2 Announce Type: replace-cross Abstract: Long video summarization presents significant challenges for multimodal large language models (MLLMs), particularly in maintaining temporal fidelity over extended durations and producing summaries that are both semantically and temporally grounded.
The paper introduces OACap, a caption-guided active observation system for detailed video captioning that requests targeted observations to fill missing information and iteratively refines its captions. OACap consistently improves performance on general video-understanding benchmarks while using only a text‑only reader. The authors also present FSCap‑Bench, a new evaluation suite with Core‑Support Factuality Score and Cross‑Modal Semantic Round‑Trip Score to assess factual and semantic preservation in captions.
arXiv:2505.01583v2 Announce Type: replace Abstract: Understanding causal event relationships and achieving fine-grained temporal grounding in videos remain challenging for vision-language models (VLM...
PercepCap is a video captioning framework that explicitly models spatio‑temporal perception before generating captions. It follows a perceive‑describe chain, first producing a perception trace of object trajectories and temporal events, then generating the final caption conditioned on that trace. The method uses a two‑stage training strategy—supervised fine‑tuning followed by perception‑grounded reinforcement learning—and builds caption‑aligned perception data to ensure the perception trace and caption refer to the same objects and events.
Automatic evaluation of image and video captioning is essential for benchmarking multimodal systems, although standard evaluation metrics show limited alignment with human judgments. Recent approaches using large language models (LLMs), commonly referred to as LLM-as-a-Judge, have improved alignment with human judgments but still suffer from a mismatch between large-vocabulary language modeling and evaluation over a small label set.