Fine-tuning X-VLM on 10,000 image-text pairs from PAB reaches 85.49 Recall@1 on the TPAS test set, ranking fourth of six teams.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Efficient Vision Language Model Fine-tuning for Text-based Person Anomaly Search
Fine-tuning X-VLM on 10,000 image-text pairs from PAB reaches 85.49 Recall@1 on the TPAS test set, ranking fourth of six teams.