Search Results for author:
Found 51 papers, 34 papers with code
Low-bit Model Quantization for Deep Neural Networks: A Survey
Therefore, it has become increasingly popular and critical to investigate how to perform the conversion and how to compensate for the information loss.
RGB-Event Fusion with Self-Attention for Collision Prediction
Notably, the event-based model outperforms the RGB model by 4% for position and 26% for time error at a similar computational cost, making it a competitive alternative.
An Empirical Study of Qwen3 Quantization
The Qwen series has emerged as a leading family of open-source Large Language Models (LLMs), demonstrating remarkable capabilities in natural language understanding tasks.
Enhancing Autonomous Driving Systems with On-Board Deployed Large Language Models
Neural Networks (NNs) trained through supervised learning struggle with managing edge-case scenarios common in real-world driving due to the intractability of exhaustive datasets covering all edge-cases, making knowledge-driven approaches, akin to how humans intuitively detect unexpected driving behavior, a suitable complement to data-driven methods.
Q-MambaIR: Accurate Quantized Mamba for Efficient Image Restoration
While low-bit quantization is an efficient model compression strategy for reducing size and accelerating IR tasks, SSM suffers substantial performance drops at ultra-low bit-widths (2-4 bits), primarily due to outliers that exacerbate quantization error.
TR-DQ: Time-Rotation Diffusion Quantization
TR-DQ achieves state-of-the-art (SOTA) performance on image generation and video generation tasks and a 1. 38-1. 89x speedup and 1. 97-2. 58x memory reduction in inference compared to existing quantization methods.
QArtSR: Quantization via Reverse-Module and Timestep-Retraining in One-Step Diffusion based Image Super-Resolution
To explore more possibilities of quantized OSDSR, we propose an efficient method, Quantization via reverse-module and timestep-retraining for OSDSR, named QArtSR.
Q&C: When Quantization Meets Cache in Efficient Image Generation
In this work, we take advantage of these two acceleration mechanisms and propose a hybrid acceleration method by tackling the above challenges, aiming to further improve the efficiency of DiTs while maintaining excellent generation capability.
MPQ-DM: Mixed Precision Quantization for Extremely Low Bit Diffusion Models
However, the existing quantization methods for diffusion models still cause severe degradation in performance, especially under extremely low bit-widths (2-4 bit).
DynamicPAE: Generating Scene-Aware Physical Adversarial Examples in Real-Time
The key challenges in generating dynamic PAEs are exploring their patterns under noisy gradient feedback and adapting the attack to agnostic scenario natures.
BiDM: Pushing the Limit of Quantization for Diffusion Models
Diffusion models (DMs) have been significantly developed and widely used in various applications due to their excellent generative qualities.
PassionSR: Post-Training Quantization with Adaptive Scale in One-Step Diffusion based Image Super-Resolution
Diffusion-based image super-resolution (SR) models have shown superior performance at the cost of multiple denoising steps.
BiDense: Binarization for Dense Prediction
BiDense incorporates two key techniques: the Distribution-adaptive Binarizer (DAB) and the Channel-adaptive Full-precision Bypass (CFB).
GWQ: Gradient-Aware Weight Quantization for Large Language Models
Compared to current quantization methods, GWQ can be applied to multiple language models and achieves lower PPL on the WikiText2 and C4 dataset.
LLMCBench: Benchmarking Large Language Model Compression for Efficient Deployment
Finally, we perform an in-depth analysis based on the evaluation and provide useful insight for LLM compression design.
ODDN: Addressing Unpaired Data Challenges in Open-World Deepfake Detection on Online Social Networks
To overcome this issue, we propose a novel approach named the open-world deepfake detection network (ODDN), which comprises two core modules: open-world data aggregation (ODA) and compression-discard gradient correction (CGC).
ARB-LLM: Alternating Refined Binarizations for Large Language Models
However, current binarization methods struggle to narrow the distribution gap between binarized and full-precision weights, while also overlooking the column deviation in LLM weight distribution.
A Survey of Low-bit Large Language Models: Basics, Systems, and Algorithms
Large language models (LLMs) have achieved remarkable advancements in natural language processing, showcasing exceptional performance across various tasks.
2DQuant: Low-bit Post-Training Quantization for Image Super-Resolution
Low-bit quantization has become widespread for compressing image super-resolution (SR) models for edge deployment, which allows advanced SR models to enjoy compact low-bit parameters and efficient integer/bitwise constructions for storage compression and inference acceleration, respectively.
Binarized Diffusion Model for Image Super-Resolution
Nonetheless, due to the model structure and the multi-step iterative attribute of DMs, existing binarization methods result in significant performance degradation.
SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models
Specifically, the proposed SliM-LLM mainly relies on two novel techniques: (1) Salience-Determined Bit Allocation utilizes the clustering characteristics of salience distribution to allocate the bit-widths of each group, increasing the accuracy of quantized LLMs and maintaining the inference efficiency; (2) Salience-Weighted Quantizer Calibration optimizes the parameters of the quantizer by considering the element-wise salience within the group, balancing the maintenance of salient information and minimization of errors.
An empirical study of LLaMA3 quantization: from LLMs to MLLMs
To uncover the capabilities of low-bit quantized MLLM, we assessed the performance of the LLaMA3-based LLaVA-Next-8B model under 2-4 ultra-low bits with post-training quantization methods.
BinaryDM: Accurate Weight Binarization for Efficient Diffusion Models
From the optimization perspective, a Low-rank Representation Mimicking (LRM) is applied to assist the optimization of binarized DMs.
Graph Construction with Flexible Nodes for Traffic Demand Prediction
To tackle these challenges, this paper introduces a novel graph construction method tailored to free-floating traffic mode.
DB-LLM: Accurate Dual-Binarization for Efficient LLMs
Large language models (LLMs) have significantly advanced the field of natural language processing, while the expensive memory and computation consumption impede their practical deployment.
Accurate LoRA-Finetuning Quantization of LLMs via Information Retention
This paper proposes a novel IR-QLoRA for pushing quantized LLMs with LoRA to be highly accurate through information retention.
BiLLM: Pushing the Limit of Post-Training Quantization for LLMs
Pretrained large language models (LLMs) exhibit exceptional general language processing capabilities but come with significant demands on memory and computational resources.
Image Fusion via Vision-Language Model
Therefore, we introduce a novel fusion paradigm named image Fusion via vIsion-Language Model (FILM), for the first time, utilizing explicit textual information from source images to guide the fusion process.
RdimKD: Generic Distillation Paradigm by Dimensionality Reduction
Knowledge Distillation (KD) emerges as one of the most promising compression technologies to run advanced deep neural networks on resource-limited devices.
BinaryHPE: 3D Human Pose and Shape Estimation via Binarization
In this work, we propose BinaryHPE, a novel binarization method designed to estimate the 3D human body, face, and hands parameters efficiently.
On-Chip Hardware-Aware Quantization for Mixed Precision Neural Networks
This leads to a non-negligible gap between the estimated efficiency metrics and the actual hardware that makes quantized models far away from the optimal accuracy and efficiency, and also causes the quantization process to rely on additional high-performance devices.
RobustMQ: Benchmarking Robustness of Quantized Models
Quantization has emerged as an essential technique for deploying deep neural networks (DNNs) on devices with limited resources.
How Good is Google Bard's Visual Understanding? An Empirical Study on Open Challenges
Google's Bard has emerged as a formidable competitor to OpenAI's ChatGPT in the field of conversational AI.
Benchmarking the Robustness of Quantized Models
To address this gap, we thoroughly evaluated the robustness of quantized models against various noises (adversarial attacks, natural corruptions, and systematic noises) on ImageNet.
Towards Accurate Post-Training Quantization for Vision Transformer
We find the main reasons lie in (1) the existing calibration metric is inaccurate in measuring the quantization influence for extremely low-bit representation, and (2) the existing quantization paradigm is unfriendly to the power-law distribution of Softmax.
BiBench: Benchmarking and Analyzing Network Binarization
Network binarization emerges as one of the most promising compression approaches offering extraordinary computation and memory savings by minimizing the bit-width.
BiFSMNv2: Pushing Binary Neural Networks for Keyword Spotting to Real-Network Performance
We highlight that benefiting from the compact architecture and optimized hardware kernel, BiFSMNv2 can achieve an impressive 25. 1x speedup and 20. 2x storage-saving on edge hardware.
Defensive Patches for Robust Recognition in the Physical World
For the generalization against diverse noises, we inject class-specific identifiable patterns into a confined local patch prior, so that defensive patches could preserve more recognizable features towards specific classes, leading models for better recognition under noises.
BiBERT: Accurate Fully Binarized BERT
The large pre-trained BERT has achieved remarkable performance on Natural Language Processing (NLP) tasks but is also computation and memory expensive.
BiFSMN: Binary Neural Network for Keyword Spotting
Then, to allow the instant and adaptive accuracy-efficiency trade-offs at runtime, we also propose a Thinnable Binarization Architecture to further liberate the acceleration potential of the binarized network from the topology perspective.
Hardware-friendly Deep Learning by Network Quantization and Binarization
Quantization is emerging as an efficient approach to promote hardware-friendly deep learning and run deep neural networks on resource-limited hardware.
Distribution-sensitive Information Retention for Accurate Binary Neural Network
We present a novel Distribution-sensitive Information Retention Network (DIR-Net) that retains the information in the forward and backward propagation by improving internal propagation and introducing external representations.
Diverse Sample Generation: Pushing the Limit of Generative Data-free Quantization
We first give a theoretical analysis that the diversity of synthetic samples is crucial for the data-free quantization, while in existing approaches, the synthetic data completely constrained by BN statistics experimentally exhibit severe homogenization at distribution and sample levels.
Towards Real-world X-ray Security Inspection: A High-Quality Benchmark and Lateral Inhibition Module for Prohibited Items Detection
In this work, we first present a High-quality X-ray (HiXray) security inspection image dataset, which contains 102, 928 common prohibited items of 8 categories.
Multi-Pretext Attention Network for Few-shot Learning with Self-supervision
Self-supervised learning is emerged as an efficient method to utilize unlabeled data.
Diversifying Sample Generation for Accurate Data-Free Quantization
Unfortunately, we find that in practice, the synthetic data identically constrained by BN statistics suffers serious homogenization at both distribution level and sample level and further causes a significant performance drop of the quantized model.
Over-sampling De-occlusion Attention Network for Prohibited Items Detection in Noisy X-ray Images
The images are gathered from an airport and these prohibited items are annotated manually by professional inspectors, which can be used as a benchmark for model training and further facilitate future research.
Towards Defending Multiple $\ell_p$-norm Bounded Adversarial Perturbations via Gated Batch Normalization
In this paper, we observe that different $\ell_p$ bounded adversarial perturbations induce different statistical properties that can be separated and characterized by the statistics of Batch Normalization (BN).
BiPointNet: Binary Neural Network for Point Clouds
To alleviate the resource constraint for real-time point cloud applications that run on edge devices, in this paper we present BiPointNet, the first model binarization approach for efficient deep learning on point clouds.
Binary Neural Networks: A Survey
The binary neural network, largely saving the storage and computation, serves as a promising technique for deploying deep models on resource-limited devices.
Forward and Backward Information Retention for Accurate Binary Neural Networks
Our empirical study indicates that the quantization brings information loss in both forward and backward propagation, which is the bottleneck of training accurate binary neural networks.
