Reward-guided hyperparameter search turns SAM into a task-specific microscope-particle segmenter, with validation limited to single images and no ground truth.
Unsupervised template learning for fine-grained object recognition
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
-
SAM$^{*}$: Task-Adaptive SAM with Physics-Guided Rewards
Reward-guided hyperparameter search turns SAM into a task-specific microscope-particle segmenter, with validation limited to single images and no ground truth.