- 전체보기(55)
- machine learning(34)
- NLP(18)
- CV(9)
- audio(8)
- CLEAN CODE(7)
- hackathon(5)
- algorithm(2)
- EdgeTPU(1)
Not Only Rewards But Also Constraints: Applications on Legged Robot Locomotion
Paper: https://arxiv.org/abs/2308.12517
AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control
Paper: https://arxiv.org/abs/2104.02180
MoAI: Mixture of All Intelligence for Large Language and Vision Models
Paper : https://arxiv.org/abs/2403.07508
Mixture of Attention Heads: Selecting Attention Heads Per Token
Paper: https://arxiv.org/abs/2210.05144
FedBABU: Towards Enhanced Representation for Federated Image Classification
Paper : https://arxiv.org/abs/2106.06042
Federated Dynamic Sparse Training: Computing Less, Communicating Less, Yet Learning Better
Paper : https://arxiv.org/abs/2112.09824
Sparse Model Soups: A recipe for Improved Pruning via Model Averaging
Paper : https://arxiv.org/abs/2306.16788
Complement Sparsification: Low-Overhead Model Pruning for Federated Learning
Paper : https://arxiv.org/abs/2303.06237
Audio Deepfake Detection : A Survey
Paper : https://arxiv.org/abs/2308.14970
[Capstone #10] Encoding Speaker-Specific Latent Speech Feature for Speech Synthesis
Paper : https://arxiv.org/abs/2311.11745
[Capstone #9] Sound Design Strategies for Latent Audio Space Explorations Using Deep Learning Architectures
Paper : https://arxiv.org/abs/2305.15571
DLT: Conditioned layout generation with Joint Discrete-Continuous Diffusion Layout Transformer
Paper : https://arxiv.org/abs/2303.03755
[UROP #14] Exploring Gender Bias in Pre-Trained Transformer-based Language Models Using Movement Pruning
Paper : https://arxiv.org/abs/2207.02463
[Capstone #8] SK TECH SUMMIT 2023: VITS2
Youtube : https://www.youtube.com/watch?v=Abov0q9T4jU:
[Capstone #7] Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech
Paper : https://arxiv.org/abs/2106.06103
[UROP #13] Block Pruning For Faster Transformers
Paper : https://arxiv.org/abs/2109.04838
[Capstone #6] VITS2: Improving Quality and Efficiency of Single-Stage Text-to-Speech with Adversarial Learning and Architecture Design
Paper : https://arxiv.org/abs/2307.16430
[Capstone #5] Matcha-TTS: A fast TTS architecture with conditional flow matching
Paper : https://arxiv.org/abs/2309.03199
[Capstone #4] OverFlow: Putting flows on top of neural transducers for better TTS
Paper : https://arxiv.org/abs/2211.06892