Social Media Robustness Bench: SDXL+InstantID Image Detection

A benchmark dataset for evaluating the robustness of image detectors against compression artifacts from major social media platforms. The description indicates it tests detector performance under compression from Instagram, Facebook, TikTok, and X. It is licensed under CC BY-NC 4.0.

Use Cases

• Benchmarking AI-generated image detector robustness based on social media compression mentioned in the description
• Studying the impact of platform-specific compression algorithms on detection accuracy
• Training or fine-tuning detection models to be resilient to image artifacts from Instagram, Facebook, TikTok, and X

Strengths

• Focuses on a practical challenge: detector robustness under real-world social media compression.
• Explicitly tests against compression from four major platforms: Instagram, Facebook, TikTok, and X.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Collection Method

Likely contains images processed through social media compression pipelines for benchmarking.