arXiv Computer Vision
When the Forger Is the Judge: GPT-Image-2 Cannot Recognize Its Own Faked Documents
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
4d ago
AIForge-Doc: A Benchmark for Detecting AI-Forged Tampering in Financial and Form Documents
arXiv:2602.20569v2 Announce Type: replace Abstract: We present AIForge-Doc, the first dedicated benchmark targeting exclusively diffusion-model-based inpainting in financial and form documents with p...
arXiv Computer Vision
4d ago
DOCFORGE-BENCH: A Comprehensive 0-shot Benchmark for Document Forgery Detection and Analysis
arXiv:2603.01433v3 Announce Type: replace Abstract: We present DOCFORGE-BENCH, the first unified zero-shot benchmark for document forgery detection, evaluating 14 methods across eight datasets spanni...
arXiv Computer Vision
Sep 15
From Advertised Improvements to Measured Capabilities: Evaluating ChatGPT Images 2.5 on Forgery Tasks
arXiv:2609.13617v1 Announce Type: new Abstract: We evaluate whether the improvements advertised for ChatGPT Images 2.5 translate into better performance on forgery tasks with predetermined answers. W...
arXiv Computer Vision
Sep 17
ChatGPT Images 2.5 on Forgery Tasks: Testing Advertised Improvements Against Known Answers
arXiv:2609.13617v2 Announce Type: replace Abstract: OpenAI released ChatGPT Images 2.5 on 8 September 2026, advertising more precise local edits, better consistency across edits, more faithful refere...
arXiv AI
4d ago
GPT4o-Receipt: A Dataset and Human Study for AI-Generated Document Forensics
GPT4o-Receipt is a benchmark dataset of 1,235 receipt images that pairs GPT‑4o‑generated receipts with authentic ones from established datasets. The study evaluates five state‑of‑the‑art multimodal LLMs and a 30‑annotator crowdsourced perceptual study, revealing that humans are better at spotting AI artifacts yet perform worse at binary detection than models like Claude Sonnet 4 and Gemini 2.5 Flash. The paradox is explained by the fact that the dominant forensic signals—arithmetic errors—are invisible to visual inspection but can be quickly verified by LLMs.
arXiv Computer Vision
Oct 16
Beyond Generation: Using Multimodal LLMs as Detectors of LLM-Generated Deepfake Facial Images
The paper evaluates whether multimodal large language models (MLLMs) can serve as forensic detectors of deepfake facial images. Six MLLMs—including GPT-4o, GPT-5, GPT-5.2, Gemini 2.5, Gemini 3, and Seed 1.6—were tested on a dataset of 400 images (50 authentic and 350 synthetic) generated by seven leading image models. Results show that several MLLMs achieve high detection accuracy, with GPT-5.2 reaching 97.9% in pairwise comparisons, but the presence of invisible SynthID watermarks can reduce detection performance by about 10% for Gemini.