Haotian Liu

7

Papers

187

Total Citations

Papers (7)

ViP-LLaVA: Making Large Multimodal Models Understand Arbitrary Visual Prompts

SeaS: Few-shot Industrial Anomaly Image Generation with Separation and Sharing Fine-tuning

Removing Distributional Discrepancies in Captions Improves Image-Text Alignment

AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios

Edit One for All: Interactive Batch Image Editing

Generalizable Face Landmarking Guided by Conditional Face Warping

Improved Baselines with Visual Instruction Tuning