Picture for Simyung Chang

Simyung Chang

Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP

Add code
Oct 11, 2024
Figure 1 for Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP
Figure 2 for Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP
Figure 3 for Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP
Figure 4 for Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP
Viaarxiv icon

InfiniPot: Infinite Context Processing on Memory-Constrained LLMs

Add code
Oct 02, 2024
Viaarxiv icon

Feature Diversification and Adaptation for Federated Domain Generalization

Add code
Jul 11, 2024
Figure 1 for Feature Diversification and Adaptation for Federated Domain Generalization
Figure 2 for Feature Diversification and Adaptation for Federated Domain Generalization
Figure 3 for Feature Diversification and Adaptation for Federated Domain Generalization
Figure 4 for Feature Diversification and Adaptation for Federated Domain Generalization
Viaarxiv icon

Crayon: Customized On-Device LLM via Instant Adapter Blending and Edge-Server Hybrid Inference

Add code
Jun 11, 2024
Viaarxiv icon

Improving Small Footprint Few-shot Keyword Spotting with Supervision on Auxiliary Data

Add code
Aug 31, 2023
Figure 1 for Improving Small Footprint Few-shot Keyword Spotting with Supervision on Auxiliary Data
Figure 2 for Improving Small Footprint Few-shot Keyword Spotting with Supervision on Auxiliary Data
Figure 3 for Improving Small Footprint Few-shot Keyword Spotting with Supervision on Auxiliary Data
Figure 4 for Improving Small Footprint Few-shot Keyword Spotting with Supervision on Auxiliary Data
Viaarxiv icon

Knowledge Distillation from Non-streaming to Streaming ASR Encoder using Auxiliary Non-streaming Layer

Add code
Aug 31, 2023
Viaarxiv icon

Scalable Weight Reparametrization for Efficient Transfer Learning

Add code
Feb 26, 2023
Viaarxiv icon

Quadapter: Adapter for GPT-2 Quantization

Add code
Nov 30, 2022
Figure 1 for Quadapter: Adapter for GPT-2 Quantization
Figure 2 for Quadapter: Adapter for GPT-2 Quantization
Figure 3 for Quadapter: Adapter for GPT-2 Quantization
Figure 4 for Quadapter: Adapter for GPT-2 Quantization
Viaarxiv icon

QTI Submission to DCASE 2021: residual normalization for device-imbalanced acoustic scene classification with efficient design

Add code
Jun 28, 2022
Figure 1 for QTI Submission to DCASE 2021: residual normalization for device-imbalanced acoustic scene classification with efficient design
Figure 2 for QTI Submission to DCASE 2021: residual normalization for device-imbalanced acoustic scene classification with efficient design
Figure 3 for QTI Submission to DCASE 2021: residual normalization for device-imbalanced acoustic scene classification with efficient design
Figure 4 for QTI Submission to DCASE 2021: residual normalization for device-imbalanced acoustic scene classification with efficient design
Viaarxiv icon

Personalized Keyword Spotting through Multi-task Learning

Add code
Jun 28, 2022
Figure 1 for Personalized Keyword Spotting through Multi-task Learning
Figure 2 for Personalized Keyword Spotting through Multi-task Learning
Figure 3 for Personalized Keyword Spotting through Multi-task Learning
Figure 4 for Personalized Keyword Spotting through Multi-task Learning
Viaarxiv icon