Search Results for author:
Found 16 papers, 6 papers with code
IG-FIQA: Improving Face Image Quality Assessment through Intra-class Variance Guidance robust to Inaccurate Pseudo-Labels
In the realm of face image quality assesment (FIQA), method based on sample relative classification have shown impressive performance.
VIGFace: Virtual Identity Generation Model for Face Image Synthesis
Initially, we train the face recognition model using a real face dataset and create a feature space for both real and virtual IDs where virtual prototypes are orthogonal to other prototypes.
Enhancing Computation Efficiency in Large Language Models through Weight and Activation Quantization
Large Language Models (LLMs) are proficient in natural language processing tasks, but their deployment is often restricted by extensive parameter sizes and computational demands.
Token-Scaled Logit Distillation for Ternary Weight Generative Language Models
Generative Language Models (GLMs) have shown impressive performance in tasks such as text generation, understanding, and reasoning.
Self-supervised Equality Embedded Deep Lagrange Dual for Approximate Constrained Optimization
Furthermore, we prove the convergence of DeepLDE and show that the primal-dual learning method alone cannot ensure equality constraints without the help of equality embedding.
Meta-Explore: Exploratory Hierarchical Vision-and-Language Navigation Using Scene Object Spectrum Grounding
We show that an exploitation policy, which moves the agent toward a well-chosen local goal among unvisited but observable states, outperforms a method which moves the agent to a previously visited state.
Ranked #1 on
Visual Navigation
on R2R
Teacher Intervention: Improving Convergence of Quantization Aware Training for Ultra-Low Precision Transformers
Pre-trained Transformer models such as BERT have shown great success in a wide range of applications, but at the cost of substantial increases in model complexity.
Understanding and Improving Knowledge Distillation for Quantization-Aware Training of Large Transformer Encoders
In particular, KD has been employed in quantization-aware training (QAT) of Transformer encoders like BERT to improve the accuracy of the student model with the reduced-precision weight parameters.
Privacy-Preserving Text Classification on BERT Embeddings with Homomorphic Encryption
Embeddings, which compress information in raw text into semantics-preserving low-dimensional vectors, have been widely adopted for their efficacy.
SR-GCL: Session-Based Recommendation with Global Context Enhanced Augmentation in Contrastive Learning
As a crucial component of contrastive learning, we propose two global context enhanced data augmentation methods while maintaining the semantics of the original session.
Collective Relevance Labeling for Passage Retrieval
Deep learning for Information Retrieval (IR) requires a large amount of high-quality query-document relevance labels, but such labels are inherently sparse.
NN-LUT: Neural Approximation of Non-Linear Operations for Efficient Transformer Inference
Non-linear operations such as GELU, Layer normalization, and Softmax are essential yet costly building blocks of Transformer models.
Deep learning-based statistical noise reduction for multidimensional spectral data
In such case, the limited time available for data acquisition can be a serious constraint for experiments in which multidimensional spectral data are acquired.
A 3D model-based approach for fitting masks to faces in the wild
Our method proceeds by first fitting a 3D morphable model on the input image, second overlaying the mask surface onto the face model and warping the respective mask texture, and last projecting the 3D mask back to 2D.
A Unified Framework for Tumor Proliferation Score Prediction in Breast Histopathology
We present a unified framework to predict tumor proliferation scores from breast histopathology whole slide images.
Towards Abstraction from Extraction: Multiple Timescale Gated Recurrent Unit for Summarization
The results also show that the temporal hierarchies help improve the ability of seq2seq models to capture compositionalities better without the presence of highly complex architectural hierarchies.
