FashionLens is a task-adaptive MLLM framework that achieves SOTA performance on diverse fashion image retrieval scenarios via spherical query calibration and gradient-guided sampling.
BLIP: bootstrapping language-image pre-training for unified vision-language understanding and generation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
FashionLens: Toward Versatile Fashion Image Retrieval via Task-Adaptive Learning
FashionLens is a task-adaptive MLLM framework that achieves SOTA performance on diverse fashion image retrieval scenarios via spherical query calibration and gradient-guided sampling.