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...
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.
VTR-Bench is a new benchmark designed to evaluate how well video generation models render text within scenes. It includes 300 prompts across five real-world scenarios such as advertisements and scientific videos, and uses an automated pipeline with human alignment to assess text fidelity and scene/motion requirements. Experiments on 11 state‑of‑the‑art models show that even the best performer has a word error rate of 0.250, underscoring widespread challenges in visual text rendering.
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.
The paper introduces PEBench, a unified benchmark for evaluating prompt enhancers (PEs) in video generation across text-to-video, image-to-video, and reference-to-video tasks. It contains 1,100 expert-verified cases, 1,005 visual assets, and 35 fine-grained tasks, and employs a fact-extraction and rubric-based assessment framework covering 24 criteria. Evaluation of various PE methods shows a trend toward intent-preserving cinematic planning, with certain methods excelling in coverage or coherence, and human validation confirms that PEBench scores align with expert judgments and downstream video quality.
The paper introduces Redemption Score (RS), a multi‑modal evaluation framework for image captioning that combines three complementary signals: Mutual Information Divergence for global image‑text alignment, DINO‑based perceptual similarity of cycle‑generated images for visual grounding, and LLM text embeddings for contextual similarity to human references. RS fuses these signals to provide a more holistic assessment, achieving a Kendall‑τ of 58.42 on Flickr8k and outperforming most prior methods. The framework demonstrates consistent performance across Conceptual Captions and MS COCO, offering a robust evaluation that captures both visual accuracy and text quality.
Video captioning requires fine-grained spatio-temporal understanding of videos, including spatial perception of where objects are located and temporal perception of when events occur. Existing MLLMs usually generate captions directly from video inputs without exposing the perceptual evidence behind descriptions.
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.
SVG-Score introduces a human‑aligned evaluation framework for text‑to‑SVG generation, addressing the shortcomings of existing image‑based metrics like CLIPScore that poorly capture SVG‑specific errors such as color, count, and spatial inaccuracies. The authors first demonstrate that CLIP‑based scores are largely insensitive to these errors and that generic Vision‑Language Models respond inconsistently across error types and styles. They then present a human‑annotated Semantic Alignment dataset and develop two complementary evaluators: a CLIP‑based scorer adapted to vector graphics and a VLM judge refined through supervised fine‑tuning and reward‑shaped reinforcement learning, enabling both fast large‑scale and expressive, interpretable assessment of SVG generators.
PANORAMA introduces a new panoptic grounded captioning framework that jointly generates detailed image captions and associates each phrase with precise pixel-level masks. The authors create PanoCaps, a human‑annotated benchmark with dense captions and near‑complete pixel coverage, and propose a phrase‑mask matching protocol with a generalized Panoptic Quality metric. PANORAMA conditions a pretrained segmenter on contextualized phrase representations, learns to select appropriate masks, and achieves state‑of‑the‑art grounding performance on PanoCaps and other pixel‑level tasks.
MTOR introduces a generalizable AI‑generated video detector that combines global visual features with caption‑derived textual semantics and a novel Temporal Over‑Regularity (TOR) component. The TOR module captures three levels of temporal consistency—coarse inter‑frame continuity, fine‑grained token correspondence, and frame‑to‑video stability—to exploit the stronger temporal persistence and lower variability found in synthetic videos. Extensive tests on five benchmarks with 46 generator variants show MTOR outperforms 16 baselines and remains robust against twelve real‑world video perturbations.
ClearText-Video (CTVid) is a large-scale, scene-text-aware benchmark that examines text-centric video understanding under varying quality conditions. It comprises 4,639 real-world egocentric videos, over 550,000 frames, 1.6 million human-verified scene-text annotations, and more than 220,000 spatial/temporal question–answer pairs in Chinese and English. For each high-quality video, CTVid provides matched degraded- and restored-quality variants, enabling studies of Text-Centric Video Restoration and Multi-Quality VideoQA, and revealing that visual enhancement does not always improve textual fidelity or downstream reasoning.