Semantic-Aware Contrasting with Explicitly Disentangled Representations for AI-Generated Video Detection

The rapid proliferation of AI-generated videos (AIGVs) has raised critical concerns about content authenticity and trustworthy media consumption, underscoring the need for robust AIGV detection. Despite recent progress, existing benchmarks suffer from a fundamental limitation: real and fake videos are often collected or generated from different semantic distributions. As a result, detectors may rely on spurious correlations involving objects, scenes, or actions rather than intrinsic artifacts introduced by generative models. This limitation calls for a benchmark that explicitly aligns real and generated videos at the semantic level, enabling detection methods to be evaluated and optimized under controlled content conditions. To this end, we introduce SAVGE, a large-scale semantically aligned AIGV detection dataset comprising 2,369,822 videos, including 424,115 real videos and 1,945,707 generated videos. SAVGE organizes samples into real--fake semantic groups by using captions paired with real videos as prompts for text-to-video generation, thereby providing both group-level correspondence and textual semantic anchors. Based on SAVGE, we further propose Explicit Disentangling and Semantic-aware Contrasting (EDSC), a model-agnostic training framework for learning artifact-focused representations with reduced semantic interference. EDSC first extracts and aligns visual semantic features with textual representations, and then explicitly suppresses semantic information in the artifact branch while contrasting real and fake samples within each semantic group. By integrating caption-guided semantic supervision with intra-group contrastive learning in a two-phase optimization strategy, EDSC encourages detectors to capture generation-related cues under matched semantic conditions. Extensive experiments on SAVGE and GenVideo demonstrate that EDSC improves most evaluated backbone--metric combinations, enhances robustness under challenging evaluation settings, and strengthens cross-dataset generalization. These results highlight the importance of semantic alignment in AIGV detection and validate explicit semantic--artifact disentanglement as an effective strategy for robust detector learning.

Authors

Institutions

Publication Details

Journal
Journal of Intelligent Computing and Networking
Published
2026-11-16
DOI
https://doi.org/10.64509/jicn.24.159
Primary Topic
Digital Media Forensic Detection
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Semantic-Aware Contrasting with Explicitly Disentangled Representations for AI-Generated Video Detection

Jiawei Liu, Yazhuo Qian, Zeyu Zhang, Jiaquan Yu et al.
1 links
Journal of Intelligent Computing and Networking
Digital Media Forensic Detection
article

Semantic-Aware Contrasting with Explicitly Disentangled Representations for AI-Generated Video Detection

Jiawei Liu, Yazhuo Qian, Zeyu Zhang, Jiaquan Yu, Jingyi Xie, Xiaobin Hu
article en

Abstract

The rapid proliferation of AI-generated videos (AIGVs) has raised critical concerns about content authenticity and trustworthy media consumption, underscoring the need for robust AIGV detection. Despite recent progress, existing benchmarks suffer from a fundamental limitation: real and fake videos are often collected or generated from different semantic distributions. As a result, detectors may rely on spurious correlations involving objects, scenes, or actions rather than intrinsic artifacts introduced by generative models. This limitation calls for a benchmark that explicitly aligns real and generated videos at the semantic level, enabling detection methods to be evaluated and optimized under controlled content conditions. To this end, we introduce SAVGE, a large-scale semantically aligned AIGV detection dataset comprising 2,369,822 videos, including 424,115 real videos and 1,945,707 generated videos. SAVGE organizes samples into real--fake semantic groups by using captions paired with real videos as prompts for text-to-video generation, thereby providing both group-level correspondence and textual semantic anchors. Based on SAVGE, we further propose Explicit Disentangling and Semantic-aware Contrasting (EDSC), a model-agnostic training framework for learning artifact-focused representations with reduced semantic interference. EDSC first extracts and aligns visual semantic features with textual representations, and then explicitly suppresses semantic information in the artifact branch while contrasting real and fake samples within each semantic group. By integrating caption-guided semantic supervision with intra-group contrastive learning in a two-phase optimization strategy, EDSC encourages detectors to capture generation-related cues under matched semantic conditions. Extensive experiments on SAVGE and GenVideo demonstrate that EDSC improves most evaluated backbone--metric combinations, enhances robustness under challenging evaluation settings, and strengthens cross-dataset generalization. These results highlight the importance of semantic alignment in AIGV detection and validate explicit semantic--artifact disentanglement as an effective strategy for robust detector learning.

Journal of Intelligent Computing and NetworkingVol. 2(4)
University of Science and Technology of China (CN), National University of Singapore (SG)
National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities
Digital Media Forensic Detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon