Research map / Vision-language and multimodal / Multimodal LLMs
CogniCred: A Dataset and Benchmark for Cognitive Credential Forgery Detection
Junchi Li, Jiasheng Sun, Weizhi Chen, Ziwei Wang, Sheng Zhou, Jiajun Bu, Chenfan Qu, Bohan Yu, Jian liu, Weiqiang Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| enable Thinking mode | Qwen3-VL-32B · CogniCredBench Contextual Reasoning (T1 closed-set) | accuracy (%): 56.85 -> 85.80 | Improved |
| enable Thinking mode | Qwen3-VL-30B-A3B · CogniCredBench Contextual Reasoning (T1 closed-set) | accuracy (%): 64.45 -> 75.25 | Improved |
| enable Thinking mode | Qwen3-VL-32B · CogniCredBench Visual Reasoning (T1 closed-set) | accuracy (%): 39.54 -> 26.99 | Hurt |
| enable Thinking mode | Qwen3-VL-30B-A3B · CogniCredBench Visual Reasoning (T1 closed-set) | accuracy (%): 26.56 -> 20.03 | Hurt |
| switch from closed-set classification (T1) to open-ended reasoning (T2) | GLM-4.5V · CogniCredBench Visual Reasoning | accuracy (%): 32.61 -> 7.74 | Hurt |
| switch from closed-set classification (T1) to open-ended reasoning (T2) | GLM-4.5V · CogniCredBench Format Compliance | accuracy (%): 29.48 -> 5.56 | Hurt |
| switch from closed-set classification (T1) to open-ended reasoning (T2) | Gemini-3-Pro · CogniCredBench Visual Reasoning | accuracy (%): 65.09 -> 44.54 | Hurt |
| switch from closed-set classification (T1) to open-ended reasoning (T2) | Qwen3-VL-32B-Instruct · CogniCredBench Visual Reasoning | accuracy (%): 39.54 -> 29.46 | Hurt |
| compare Gemini-3-Pro with GPT-4o as closed-source model | GPT-4o · CogniCredBench Visual Reasoning (T2 open-ended) | accuracy (%): 44.54 -> 1.25 | Hurt |
| enable Thinking mode | Qwen3-VL-32B · CogniCredBench Format Compliance (T2 open-ended) | accuracy (%): 36.53 -> 42.12 | Improved |
Closest papers by meaning
- Forensics-Bench: A Comprehensive Forgery Detection Benchmark Suite for Large Vision Language Models (CVPR 2025 · similarity 0.88)
- Identity-Aware Vision-Language Model for Explainable Face Forgery Detection (AAAI 2026 · similarity 0.87)
- Towards General Visual-Linguistic Face Forgery Detection (CVPR 2025 · similarity 0.86)
- A Hitchhiker's Guide to Fine-Grained Face Forgery Detection Using Common Sense Reasoning (NeurIPS 2024 · similarity 0.85)
- THEMIS: Towards Holistic Evaluation of MLLMs for Scientific Paper Fraud Forensics (ICLR 2026 · similarity 0.85)
- OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL (ICML 2026 · similarity 0.84)
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.
