SAMPLe adds dual gradient constraints (ERM alignment plus full-batch orthogonality) to SAM-style prompt learning and raises harmonic-mean base-to-new accuracy across CoOp, CoCoOp, MaPLe, TCP and CoPrompt.
In: International conference on machine learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2representative citing papers
A simple 1D-CNN video summarizer trained with new entropy-ratio and Wasserstein representativeness losses reports state-of-the-art ranking-correlation results among unsupervised methods on SumMe and TVSUM.
citing papers explorer
-
SAMPLe: SAM-based Optimizer for Prompt Learning in VLMs
SAMPLe adds dual gradient constraints (ERM alignment plus full-batch orthogonality) to SAM-style prompt learning and raises harmonic-mean base-to-new accuracy across CoOp, CoCoOp, MaPLe, TCP and CoPrompt.
-
TRIM: A Self-Supervised Video Summarization Framework Maximizing Temporal Relative Information and Representativeness
A simple 1D-CNN video summarizer trained with new entropy-ratio and Wasserstein representativeness losses reports state-of-the-art ranking-correlation results among unsupervised methods on SumMe and TVSUM.