作者位次 #13/21ACL 2026Poster· Natural Language Processing
AEGIS : A Holistic Benchmark for Evaluating Forensic Analysis of AI -Generated Academic Images
Bo Zhang · Tzu-Yen Ma · Zichen Tang · Junpeng Ding · Zirui Wang · Yizhuo Zhao · Peilin Gao · Zijie Xi · Zixin Ding · Haiyang Sun · Haocheng Gao · Yuan Liu · Liangjia Wang · Yiling Huang · Yujie Wang · Yuyue Zhang · Ronghui Xi · Yuanze Li · Jiacheng Liu · Zhongjun Yang · Haihong E
摘要
We introduce AEGIS, A holistic benchmark for Evaluating forensic analysis of AI-Generated academic ImageS. Compared to existing benchmarks, AEGIS features three key advances: (1) Domain-Specific Complexity: covering seven academic categories with 39 fine-grained subtypes, exposing intrinsic foren...