Search Results for author:
Found 13 papers, 9 papers with code
Beyond Uniform Scaling: Exploring Depth Heterogeneity in Neural Architectures
Our method is flexible towards skip connections a mainstay in modern vision transformers.
Faster and Lighter LLMs: A Survey on Current Challenges and Way Forward
Despite the impressive performance of LLMs, their widespread adoption faces challenges due to substantial computational and memory requirements during inference.
Rethinking Compression: Reduced Order Modelling of Latent Features in Large Language Models
Due to the substantial scale of Large Language Models (LLMs), the direct application of conventional compression methodologies proves impractical.
One-for-All: Generalized LoRA for Parameter-Efficient Fine-tuning
We present Generalized LoRA (GLoRA), an advanced approach for universal parameter-efficient fine-tuning tasks.
Patch Gradient Descent: Training Neural Networks on Very Large Images
Traditional CNN models are trained and tested on relatively low resolution images (<300 px), and cannot be directly operated on large-scale images due to compute and memory constraints.
On Designing Light-Weight Object Trackers through Network Pruning: Use CNNs or Transformers?
A comparison between SOTA trackers using CNNs, transformers as well as the combination of the two is presented to study their stability at various compression ratios.
Dynamic Kernel Selection for Improved Generalization and Memory Efficiency in Meta-learning
MetaDOCK compresses the meta-model as well as the task-specific inner models, thus providing significant reduction in model size for each task, and through constraining the number of active kernels for every task, it implicitly mitigates the issue of meta-overfitting.
Vision Transformer Slimming: Multi-Dimension Searching in Continuous Optimization Space
This paper explores the feasibility of finding an optimal sub-model from a vision transformer and introduces a pure vision transformer slimming (ViT-Slim) framework.
Transfer Learning Gaussian Anomaly Detection by Fine-tuning Representations
In our work, we propose a new method to overcome catastrophic forgetting and thus successfully fine-tune pre-trained representations for AD in the transfer learning setting.
ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations
Structured pruning methods are among the effective strategies for extracting small resource-efficient convolutional neural networks from their dense counterparts with minimal loss in accuracy.
Rescaling CNN through Learnable Repetition of Network Parameters
We show that small base networks when rescaled, can provide performance comparable to deeper networks with as low as 6% of optimization parameters of the deeper one.
Deep learning for detection and segmentation of artefact and disease instances in gastrointestinal endoscopy
The Endoscopy Computer Vision Challenge (EndoCV) is a crowd-sourcing initiative to address eminent problems in developing reliable computer aided detection and diagnosis endoscopy systems and suggest a pathway for clinical translation of technologies.
Multi-Plateau Ensemble for Endoscopic Artefact Segmentation and Detection
Endoscopic artefact detection challenge consists of 1) Artefact detection, 2) Semantic segmentation, and 3) Out-of-sample generalisation.
