Search Results for author:
Found 33 papers, 20 papers with code
Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weights
The weights in the other group are responsible to compensate for the accuracy loss from the quantization, thus they are the ones to be re-trained.
Deep Neural Network Compression with Single and Multiple Level Quantization
In this paper, we propose two novel network quantization approaches, single-level network quantization (SLQ) for high-bit quantization and multi-level network quantization (MLQ) for extremely low-bit quantization (ternary). We are the first to consider the network quantization from both width and depth level.
Explicit Loss-Error-Aware Quantization for Low-Bit Deep Neural Networks
Through explicitly regularizing the loss perturbation and the weight approximation error in an incremental way, we show that such a new optimization method is theoretically reasonable and practically effective.
SnapQuant: A Probabilistic and Nested Parameterization for Binary Networks
During the training phase, we generate binary weights on-the-fly since what we actually maintain is the policy network, and all the binary weights are used in a burn-after-reading style.
Adversarial Robustness vs Model Compression, or Both?
Furthermore, this work studies two hypotheses about weight pruning in the conventional setting and finds that weight pruning is essential for reducing the network model size in the adversarial setting, training a small model from scratch even with inherited initialization from the large model cannot achieve both adversarial robustness and high standard accuracy.
Deeply-supervised Knowledge Synergy
Convolutional Neural Networks (CNNs) have become deeper and more complicated compared with the pioneering AlexNet.
HBONet: Harmonious Bottleneck on Two Orthogonal Dimensions
MobileNets, a class of top-performing convolutional neural network architectures in terms of accuracy and efficiency trade-off, are increasingly used in many resourceaware vision applications.
Scale Calibrated Training: Improving Generalization of Deep Networks via Scale-Specific Normalization
Experiment results show that SCT improves accuracy of single Resnet-50 on ImageNet by 1. 7% and 11. 5% accuracy when testing on image sizes of 224 and 128 respectively.
Dynamic Graph: Learning Instance-aware Connectivity for Neural Networks
To facilitate the training, we represent the network connectivity of each sample in an adjacency matrix.
Differentiable Dynamic Wirings for Neural Networks
A standard practice of deploying deep neural networks is to apply the same architecture to all the input instances.
Learning N:M Fine-grained Structured Sparse Neural Networks From Scratch
In this paper, we are the first to study training from scratch an N:M fine-grained structured sparse network, which can maintain the advantages of both unstructured fine-grained sparsity and structured coarse-grained sparsity simultaneously on specifically designed GPUs.
Incorporating Convolution Designs into Visual Transformers
Motivated by the success of Transformers in natural language processing (NLP) tasks, there emerge some attempts (e. g., ViT and DeiT) to apply Transformers to the vision domain.
Ranked #2 on
Image Classification
on Oxford-IIIT Pets
Group Fisher Pruning for Practical Network Compression
Our method can be used to prune any structures including those with coupled channels.
DominoSearch: Find layer-wise fine-grained N:M sparse schemes from dense neural networks
However, the existing N:M algorithms only address the challenge of how to train N:M sparse neural networks in a uniform fashion (i. e. every layer has the same N:M sparsity) and suffer from a significant accuracy drop for high sparsity (i. e. when sparsity > 80\%).
Pyramid Fusion Transformer for Semantic Segmentation
The recently proposed MaskFormer gives a refreshed perspective on the task of semantic segmentation: it shifts from the popular pixel-level classification paradigm to a mask-level classification method.
Group R-CNN for Weakly Semi-supervised Object Detection with Points
The core of this task is to train a point-to-box regressor on well-labeled images that can be used to predict credible bounding boxes for each point annotation.
An Algorithm-Hardware Co-Optimized Framework for Accelerating N:M Sparse Transformers
(1) From algorithm perspective, we propose a sparsity inheritance mechanism along with an inherited dynamic pruning (IDP) method to obtain a series of N:M sparse candidate Transformers rapidly.
Omni-Dimensional Dynamic Convolution
Learning a single static convolutional kernel in each convolutional layer is the common training paradigm of modern Convolutional Neural Networks (CNNs).
SparseMAE: Sparse Training Meets Masked Autoencoders
In this paper, we aim to reduce model complexity from large vision transformers pretrained by MAE with assistant of sparse training.
FeatAug-DETR: Enriching One-to-Many Matching for DETRs with Feature Augmentation
The first method is One-to-many Matching via Data Augmentation (denoted as DataAug-DETR).
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention
We present LLaMA-Adapter, a lightweight adaption method to efficiently fine-tune LLaMA into an instruction-following model.
Ranked #2 on
Music Question Answering
on MusicQA
Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement
The popularity of Contrastive Language-Image Pre-training (CLIP) has propelled its application to diverse downstream vision tasks.
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model
This strategy effectively alleviates the interference between the two tasks of image-text alignment and instruction following and achieves strong multi-modal reasoning with only a small-scale image-text and instruction dataset.
Ranked #6 on
Visual Question Answering (VQA)
on InfiMM-Eval
Instruction Following
Optical Character Recognition (OCR)
+7
Solving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification
We found that its success can be largely attributed to its powerful skills in generating and executing code, evaluating the output of code execution, and rectifying its solution when receiving unreasonable outputs.
Ranked #1 on
Math Word Problem Solving
on MATH
Efficient N:M Sparse DNN Training Using Algorithm, Architecture, and Dataflow Co-Design
At the algorithm level, a bidirectional weight pruning method, dubbed BDWP, is proposed to leverage the N:M sparsity of weights during both forward and backward passes of DNN training, which can significantly reduce the computational cost while maintaining model accuracy.
MathCoder: Seamless Code Integration in LLMs for Enhanced Mathematical Reasoning
In this paper, we present a method to fine-tune open-source language models, enabling them to use code for modeling and deriving math equations and, consequently, enhancing their mathematical reasoning abilities.
Ranked #4 on
Math Word Problem Solving
on SVAMP
(using extra training data)
RecRanker: Instruction Tuning Large Language Model as Ranker for Top-k Recommendation
Large Language Models (LLMs) have demonstrated remarkable capabilities and have been extensively deployed across various domains, including recommender systems.
NODI: Out-Of-Distribution Detection with Noise from Diffusion
Moreover, we studied the robustness of OOD methods by applying different types of image encoders.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
Integrating Large Language Models into Recommendation via Mutual Augmentation and Adaptive Aggregation
Conventional recommendation methods have achieved notable advancements by harnessing collaborative or sequential information from user behavior.
NOAH: Learning Pairwise Object Category Attentions for Image Classification
We observe that the head structures of mainstream DNNs adopt a similar feature encoding pipeline, exploiting global feature dependencies while disregarding local ones.
Not All Experts are Equal: Efficient Expert Pruning and Skipping for Mixture-of-Experts Large Language Models
Specifically, we propose, for the first time to our best knowledge, post-training approaches for task-agnostic and task-specific expert pruning and skipping of MoE LLMs, tailored to improve deployment efficiency while maintaining model performance across a wide range of tasks.
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
To this end, we introduce MathVerse, an all-around visual math benchmark designed for an equitable and in-depth evaluation of MLLMs.
