DTM-Codec achieves better reconstruction quality and intelligibility than fixed-frame-rate neural speech codecs at matched total bitrate via dynamic token masking and Path Length Equalization for variable frame rates.
arXiv preprint arXiv:2110.03860 (2021)
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
verdicts
UNVERDICTED 4representative citing papers
Pairwise scoring signals in Vision Transformer token reduction are inherently unstable due to high perturbation counts and degrade in deep layers, causing collapse, while unary signals with triage enable CATIS to retain 96.9% accuracy at 63% FLOPs reduction on ViT-Large ImageNet-1K.
Token Merging (ToMe) doubles the throughput of large Vision Transformers on images, video, and audio by merging similar tokens with a fast matching algorithm, incurring only 0.2-0.4% accuracy loss.
Replacing selected attention heads in pretrained ViTs with depthwise convolutions, identified by simple strategies and recovered via fine-tuning, delivers 17-20% inference speedup on image tasks with minimal accuracy loss.
citing papers explorer
-
DTM-Codec: Dynamic Token Masking for VFR Speech Coding with Efficient Boundary Selection
DTM-Codec achieves better reconstruction quality and intelligibility than fixed-frame-rate neural speech codecs at matched total bitrate via dynamic token masking and Path Length Equalization for variable frame rates.
-
Why Training-Free Token Reduction Collapses: The Inherent Instability of Pairwise Scoring Signals
Pairwise scoring signals in Vision Transformer token reduction are inherently unstable due to high perturbation counts and degrade in deep layers, causing collapse, while unary signals with triage enable CATIS to retain 96.9% accuracy at 63% FLOPs reduction on ViT-Large ImageNet-1K.
-
Token Merging: Your ViT But Faster
Token Merging (ToMe) doubles the throughput of large Vision Transformers on images, video, and audio by merging similar tokens with a fast matching algorithm, incurring only 0.2-0.4% accuracy loss.
-
Accelerating Vision Foundation Models with Drop-in Depthwise Convolution
Replacing selected attention heads in pretrained ViTs with depthwise convolutions, identified by simple strategies and recovered via fine-tuning, delivers 17-20% inference speedup on image tasks with minimal accuracy loss.