AI Paper Atlas

Research map / Video / Video-language understanding

Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning

Yifei Li, Wenzhao Zheng, Yanran Zhang, Runze Sun, Yu Zheng, Lei Chen, Jie Zhou, Jiwen Lu

CVPR 2026 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 16 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add chain-of-thought grounded explanation to the SFT answer (ablation)Skyra-SFT · ViF-Benchaccuracy: 54.04 -> 90.11Improved
add SFT cold-start before RL (ablation)Skyra-RL · ViF-Benchaccuracy: 50.09 -> 91.02Improved
add GRPO RL stage after SFT (ablation)Skyra-SFT · ViF-Benchaccuracy: 90.11 -> 91.02Improved
add RL stage after SFT, especially for hard I2V samples (ablation)Skyra-SFT · ViF-Benchrecall: 84.65 -> 88.35Improved
use asymmetric accuracy reward with heavier false-positive penalty (ablation)Skyra-RL · ViF-Benchaccuracy: 76.24 -> 91.02Improved
use inspection (check-block) reward instead of plain format reward (ablation)Skyra-RL · ViF-Benchaccuracy: 90.05 -> 91.02Improved
RL adaptation on 2.2K unlabeled-annotation GenVideo samples from Skyra-SFTSkyra · GenVideo (out-of-domain)accuracy: 71.78 -> 91.00Improved
add RL stage after SFT on out-of-domain GenVideo (ablation)Skyra-SFT · GenVideo (out-of-domain)accuracy: 63.98 -> 71.78Improved
apply image degradations (JPEG compression, noise, zoom) at test timeSkyra-RL · ViF-Benchaccuracy: 91.02 -> 80.80 (compression)Hurt
prompt off-the-shelf MLLMs with CoT instead of fine-tuningQwen2.5-VL, InternVL-3, GPT-4.1-mini, Gemini-2.5-flash · ViF-Benchaccuracy: below 60%No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy