TEAM_CHEN@CC-MMD 2026: Culture-Conditioned Prompting for Zero-Shot Misogynistic Meme Detection — Asymmetric Shifts and a Feminist Probe

Abstract

CC-MMD 2026 Task B evaluates cross-cultural misogynistic meme detection. We use a fixed zero-shot VLM, Qwen2.5-VL-7B-Instruct, and vary the cultural framing of the prompt. On the English test split, positive prediction rates shift from 60.2% under a generic prompt to 61.9% under an Indian prompt, 55.0% under a Chinese prompt, and 83.8% under an Indian feminist prompt. Agreement is 78.1% between the Indian cultural and feminist settings, and 69% between the Indian prompt and a similar variant on a 100-sample subset. A Chinese sanity check shows that cross-split reuse is unreliable, with only 64% agreement. We provide a reproducible zero-shot audit framework and show that feminist prompting can serve as an effective probe of intra-cultural decision sensitivity.