SinoTechIntel Academic Portal
Official PDF TranslationFrontiers of Information Technology & Electronic Engineering

TimeJudge: empowering video-LLMs as zero-shot judges for temporal consistency in video captions

Authors: Yangliu HU; Zikai SONG; Junqing YU; Yiping Phoebe CHEN; Wei YANG

DOI: 10.1631/FITEE_2500412Status: Verified Translated Edition
Sponsored AdvertisementAd Placement Area
reCAPTCHA Bot Shield Active

Preparing Secure Academic Download

Verifying human reader & generating high-resolution document...

Verifying Document Integrity15s remaining
← Back to Article
Protected by Google reCAPTCHA v3.PrivacyTerms
Sponsored ContentAdSense In-Feed Ad Slot

Key Findings in This Report

• TimeJudge introduces a zero-shot framework that recasts temporal error detection as binary question pairs, eliminating the need for task-specific fine-tuning. • TEDBench provides a rigorously constructed benchmark with 381 videos, 1524 captions, and 3048 QA pairs, covering four complexity levels for fine-grained temporal error evaluation. • Comprehensive evaluations show that TimeJudge consistently improves recall and F1-score across multiple state-of-the-art video-LLMs for temporal consistency assessment. • The approach is generalizable, scalable, and training-free, making it suitable for real-world evaluation of video captions in multimodal systems.