V2EX 起居室老虎

Feed 地址	https://mer.dev/rss.xml
网站地址	https://mer.dev/

Speed Always Wins: A Survey on Efficient Architectures for Large Language Models

起居室老虎 AndyBlocker 4 天前

Neuromorphic Principles for Efficient Large Language Models on Intel Loihi 2

起居室老虎 AndyBlocker 13 天前

Parallelizing Linear Transformers with the Delta Rule over Sequence Length

起居室老虎 AndyBlocker 16 天前

Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity

起居室老虎 AndyBlocker 18 天前

SpikingBrain-瞬息 1.0技术报告：原生国产自主可控类脑脉冲大模型

起居室老虎 AndyBlocker 26 天前

MLP Memory: Language Modeling with Retriever-pretrained External Memory

起居室老虎 AndyBlocker 47 天前

Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention

起居室老虎 AndyBlocker 58 天前

T-MAC: CPU Renaissance via Table Lookup for Low-Bit LLM Deployment on Edge

起居室老虎 AndyBlocker 96 天前

HYTE: Flexible Tiling for Sparse Accelerators via Hybrid Static-Dynamic Approaches

起居室老虎 AndyBlocker 108 天前

Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

起居室老虎 AndyBlocker 110 天前

SpikeVideoFormer: An Efficient Spike-Driven Video Transformer with Hamming Attention and O(T) Complexity

起居室老虎 AndyBlocker 116 天前

Sparse Spiking Neural Network: Exploiting Heterogeneity in Timescales for Pruning Recurrent SNN

起居室老虎 AndyBlocker 122 天前

Prosperity: Accelerating Spiking Neural Networks via Product Sparsity

起居室老虎 AndyBlocker 122 天前

Towards Scalable GPU-Accelerated SNN Training via Temporal Fusion

起居室老虎 AndyBlocker 123 天前

Recurrent Residual Module for Fast Inference in Videos

起居室老虎 AndyBlocker 124 天前

Efficient Spatially Sparse Inference for Conditional GANs and Diffusion Models

起居室老虎 AndyBlocker 124 天前

SlowFast Networks for Video Recognition

起居室老虎 AndyBlocker 137 天前

DeltaCNN: End-to-End CNN Inference of Sparse Frame Differences in Videos

起居室老虎 AndyBlocker 141 天前

Phi: Leveraging Pattern-based Hierarchical Sparsity for High-Efficiency Spiking Neural Networks

起居室老虎 AndyBlocker 143 天前

Temporal Flexibility in Spiking Neural Networks: Towards Generalization Across Time Steps and Deployment Friendliness

起居室老虎 AndyBlocker 143 天前

A Simple Framework for Contrastive Learning of Visual Representations

起居室老虎 AndyBlocker 144 天前

QKFormer: Hierarchical Spiking Transformer using Q-K Attention

起居室老虎 AndyBlocker 156 天前

Transformers without Normalization

起居室老虎 AndyBlocker 157 天前

Visualizing and Understanding the Effectiveness of BERT

起居室老虎 AndyBlocker 158 天前

One-Minute Video Generation with Test-Time Training

起居室老虎 AndyBlocker 172 天前

Evolution Strategies as a Scalable Alternative to Reinforcement Learning

起居室老虎 AndyBlocker 173 天前

SparTA: Deep-Learning Model Sparsity via Tensor-with-Sparsity-Attribute

起居室老虎 AndyBlocker 179 天前

Scalable Diffusion Models with Transformers

起居室老虎 AndyBlocker 209 天前

初探AI Infra

起居室老虎 AndyBlocker 214 天前

Efficient and Effective Methods for Mixed Precision Neural Network Quantization for Faster, Energy-efficient Inference

起居室老虎 AndyBlocker 217 天前

Towards spike-based machine intelligence with neuromorphic computing

起居室老虎 AndyBlocker 217 天前

FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

起居室老虎 AndyBlocker 217 天前

WWW: What, When, Where to Compute-in-Memory

起居室老虎 AndyBlocker 217 天前

Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference

起居室老虎 AndyBlocker 217 天前

SpikeSim: An end-to-end Compute-in-Memory Hardware Evaluation Tool for Benchmarking Spiking Neural Networks

起居室老虎 AndyBlocker 217 天前

PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU

起居室老虎 AndyBlocker 217 天前

Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication

起居室老虎 AndyBlocker 217 天前

HAWQ: Hessian Aware Quantization of Neural Networks with Mixed-Precision

起居室老虎 AndyBlocker 217 天前

Optimizing Bit-Serial Matrix Multiplication for Reconfigurable Computing

起居室老虎 AndyBlocker 217 天前

TVM: An Automated End-to-End Optimizing Compiler for Deep Learning

起居室老虎 AndyBlocker 217 天前

Roofline: An Insightful Visual Performance Model for Floating-Point Programs and Multicore Architectures

起居室老虎 AndyBlocker 217 天前

A Comprehensive Survey on Electronic Design Automation and Graph Neural Networks: Theory and Applications

起居室老虎 AndyBlocker 217 天前

A Hardware-Software Blueprint for Flexible Deep Learning Specialization

起居室老虎 AndyBlocker 217 天前

BISMO: A Scalable Bit Serial Matrix Multiplication Overlay for Reconfigurable Computing

起居室老虎 AndyBlocker 217 天前

Code Transpilation for Hardware Accelerators

起居室老虎 AndyBlocker 217 天前

Were RNNs All We Needed?

起居室老虎 AndyBlocker 217 天前

SparseRT:Accelerating Unstructured Sparsity on GPUs for Deep Learning Inference

起居室老虎 AndyBlocker 217 天前

VPRTempo: A Fast Temporally Encoded Spiking Neural Network for Visual Place Recognition

起居室老虎 AndyBlocker 217 天前

Memory-Efficient Reversible Spiking Neural Networks

起居室老虎 AndyBlocker 217 天前

SpikeMba: Multi-Modal Spiking Saliency Mamba for Tempoal Video Grounding

起居室老虎 AndyBlocker 217 天前

SNN视频流任务调研

起居室老虎 AndyBlocker 217 天前

SpikeZIP-TF: Conversion is All You Need for Transformer-based SNN

起居室老虎 AndyBlocker 217 天前

SpikingJelly: An open-source machine learning infrastructure platform for spike-based intelligence

起居室老虎 AndyBlocker 217 天前

I-LLM: Efficient Integer-Only Inference for Fully-Quantized Low-Bit Large Language Models

起居室老虎 AndyBlocker 217 天前

程序语言理论笔记

起居室老虎 AndyBlocker 217 天前

The Minimum Equivalent DNF Problem and Shortest Implicants

起居室老虎 AndyBlocker 217 天前

I-ViT: Integer-only Quantization for Efficient Vision Transformer Inference

起居室老虎 AndyBlocker 217 天前

Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition

起居室老虎 AndyBlocker 217 天前

SpikeCV: Open a Continuous Computer Vision Era

起居室老虎 AndyBlocker 217 天前

Neuromorphic computing at scale

起居室老虎 AndyBlocker 217 天前

Titans: Learning to Memorize at Test Time

起居室老虎 AndyBlocker 217 天前

Segment Anything

起居室老虎 AndyBlocker 217 天前

SDiT: Spiking Diffusion Model with Transformer

起居室老虎 AndyBlocker 217 天前

2024

起居室老虎 AndyBlocker 217 天前

ConvUNeXt:An efficient convolution neural network for medical image segmentation

起居室老虎 AndyBlocker 217 天前

Rethinking the Membrane Dynamics and Optimization Objectives of Spiking Neural Networks

起居室老虎 AndyBlocker 217 天前

ConvNext V2: Co-designing and Scaling ConvNets with Masked Autoencoders

起居室老虎 AndyBlocker 217 天前

A ConvNet for the 2020s

起居室老虎 AndyBlocker 217 天前