Search Results for author:
Found 10 papers, 3 papers with code
EdgeFusion: On-Device Text-to-Image Generation
The intensive computational burden of Stable Diffusion (SD) for text-to-image generation poses a significant hurdle for its practical application.
Sora as an AGI World Model? A Complete Survey on Text-to-Video Generation
Text-to-video generation marks a significant frontier in the rapidly evolving domain of generative AI, integrating advancements in text-to-image synthesis, video captioning, and text-guided editing.
Shortened LLaMA: A Simple Depth Pruning for Large Language Models
Structured pruning of modern large language models (LLMs) has emerged as a way of decreasing their high computational needs.
MobileSAMv2: Faster Segment Anything to Everything
The efficiency bottleneck of SegEvery with SAM, however, lies in its mask decoder because it needs to first generate numerous masks with redundant grid-search prompts and then perform filtering to obtain the final valid masks.
Automatic Network Adaptation for Ultra-Low Uniform-Precision Quantization
Uniform-precision neural network quantization has gained popularity since it simplifies densely packed arithmetic unit for high computing capability.
Uniform-Precision Neural Network Quantization via Neural Channel Expansion
Uniform-precision neural network quantization has gained popularity thanks to its simple arithmetic unit densely packed for high computing capability.
Emotional Voice Conversion using Multitask Learning with Text-to-speech
The embedding space of seq2seq-based TTS has abundant information on the text.
Deep User Identification Model with Multiple Biometrics
Most research has only focused on single modality or a single task, while the combination of input modality or tasks is yet to be investigated.
Effective Network Compression Using Simulation-Guided Iterative Pruning
Existing high-performance deep learning models require very intensive computing.
End-to-end Multimodal Emotion and Gender Recognition with Dynamic Joint Loss Weights
In our experiments, emotion and gender recognition with the proposed method yielded a lower joint loss, which is computed as the negative log-likelihood, than using static weights for joint loss.
