Apppublic

zengrh3/mic-error-analysis

sourceHugging Faceupdated 4mo agoView on Hugging Face

0likes

README.md19 linesDownload Raw Back to root

1---2title: MIC Error Analysis3emoji: 🔍4colorFrom: red5colorTo: yellow6sdk: static7pinned: false8---9 10# MIC Error Analysis — 30 cases11 12Interactive viewer for 30 sampled errors of the MIC model (Ours-SFT-GRPO) on the TARABench test splits, grouped into three failure modes:13 14- **Mode A** — Perceptually subtle / locally-plausible edits (verdict miss)15- **Mode B** — Hallucinated visual grounding (verdict right, evidence fabricated)16- **Mode C** — Misidentified entity origin (right object, wrong country/era)17 18Open `index.html` for the interactive viewer.19