AAAI 2023 accepted papers on Architectures and efficiency
128 of the 2,023 papers accepted at AAAI 2023 (6.3%) are on Architectures and efficiency, compared with 6.6% at AAAI 2022. They fall into 8 topics.
See AAAI 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
RNNs and normalization · 46 papers
- Flow-Based Robust Watermarking with Invertible Noise Layer for Black-Box Distortions (65 citations)
- AdaTask: A Task-Aware Adaptive Learning Rate Approach to Multi-Task Learning (57 citations)
- Pointerformer: Deep Reinforced Multi-Pointer Transformer for the Traveling Salesman Problem (50 citations)
- AHPA: Adaptive Horizontal Pod Autoscaling Systems on Alibaba Cloud Container Service for Kubernetes (41 citations)
- Ising-Traffic: Using Ising Machine Learning to Predict Traffic Congestion under Uncertainty (33 citations)
- Sample-Dependent Adaptive Temperature Scaling for Improved Calibration (28 citations)
- Steganography of Steganographic Networks (27 citations)
- Resilient Binary Neural Network (25 citations)
- Scaling Law for Recommendation Models: Towards General-Purpose User Representations (25 citations)
- Predictive Exit: Prediction of Fine-Grained Early Exits for Computation- and Energy-Efficient Inference (24 citations)
- SCI: A Spectrum Concentrated Implicit Neural Compression for Biomedical Data (22 citations)
- Prompting Neural Machine Translation with Translation Memories (18 citations)
and 34 more in the interactive map.
Attention and transformers · 39 papers
- AirFormer: Predicting Nationwide Air Quality in China with Transformers (170 citations)
- Head-Free Lightweight Semantic Segmentation with Linear Transformer (109 citations)
- Spatial-Spectral Transformer for Hyperspectral Image Denoising (105 citations)
- Compact Transformer Tracker with Correlative Masked Modeling (104 citations)
- ShadowFormer: Global Context Helps Shadow Removal (93 citations)
- Graphix-T5: Mixing Pre-trained Transformers with Graph-Aware Layers for Text-to-SQL Parsing (88 citations)
- Target-Aware Tracking with Long-Term Context Attention (86 citations)
- CF-ViT: A General Coarse-to-Fine Method for Vision Transformer (70 citations)
- DeMT: Deformable Mixer Transformer for Multi-Task Learning of Dense Prediction (69 citations)
- HVTSurv: Hierarchical Vision Transformer for Patient-Level Survival Prediction from Whole Slide Image (40 citations)
- Good Helper Is around You: Attention-Driven Masked Image Modeling (25 citations)
- Personalized Dialogue Generation with Persona-Adaptive Attention (25 citations)
and 27 more in the interactive map.
Pruning and efficient tuning · 13 papers
- On the Effectiveness of Parameter-Efficient Fine-Tuning (139 citations)
- FacT: Factor-Tuning for Lightweight Adaptation on Vision Transformer (98 citations)
- Parameter-Efficient Model Adaptation for Vision Transformers (71 citations)
- A Survey on Model Compression and Acceleration for Pretrained Language Models (51 citations)
- Dynamic Structure Pruning for Compressing CNNs (21 citations)
- Compressing Transformers: Features Are Low-Rank, but Weights Are Not! (19 citations)
- HALOC: Hardware-Aware Automatic Low-Rank Compression for Compact Neural Networks (15 citations)
- Balanced Column-Wise Block Pruning for Maximizing GPU Parallelism (10 citations)
- Learning Adversarially Robust Sparse Networks via Weight Reparameterization (7 citations)
- Can We Find Strong Lottery Tickets in Generative Models? (5 citations)
- Lottery Pools: Winning More by Interpolating Tickets without Increasing Training or Inference Cost (5 citations)
- Efficient Distributed Inference of Deep Neural Networks via Restructuring and Pruning (4 citations)
and 1 more in the interactive map.
Neural architecture search · 12 papers
- PINAT: A Permutation INvariance Augmented Transformer for NAS Predictor (19 citations)
- ProxyBO: Accelerating Neural Architecture Search via Bayesian Optimization with Zero-Cost Proxies (17 citations)
- Zero-Cost Operation Scoring in Differentiable Architecture Search (13 citations)
- NAS-LID: Efficient Neural Architecture Search with Local Intrinsic Dimension (12 citations)
- Learning Conflict-Noticed Architecture for Multi-Task Learning (6 citations)
- GENNAPE: Towards Generalized Neural Architecture Performance Estimators (5 citations)
- HyperJump: Accelerating HyperBand via Risk Modelling (4 citations)
- AIO-P: Expanding Neural Performance Predictors beyond Image Classification (3 citations)
- Learning from the Wisdom of Crowds: Exploiting Similar Sessions for Session Search (2 citations)
- PatchNAS: Repairing DNNs in Deployment with Patched Network Architecture Search (2 citations)
- Dynamic Ensemble of Low-Fidelity Experts: Mitigating NAS “Cold-Start” (1 citation)
- Bespoke: A Block-Level Neural Network Optimization Framework for Low-Cost Deployment
CNN design · 7 papers
- Fourier-Net: Fast Image Registration with Band-Limited Deformation (41 citations)
- Improving Dynamic HDR Imaging with Fusion Transformer (30 citations)
- Riemannian Local Mechanism for SPD Neural Networks (24 citations)
- MSDC: Exploiting Multi-State Power Consumption in Non-intrusive Load Monitoring Based on a Dual-CNN Model (23 citations)
- EffConv: Efficient Learning of Kernel Sizes for Convolution Layers of CNNs (8 citations)
- Less Is More Important: An Attention Module Guided by Probability Density Function for Convolutional Neural Networks (5 citations)
- Testing the Channels of Convolutional Neural Networks (2 citations)
Spiking networks · 6 papers
- Reducing ANN-SNN Conversion Error through Residual Membrane Potential (54 citations)
- ESL-SNNs: An Evolutionary Structure Learning Strategy for Spiking Neural Networks (49 citations)
- Exploring Temporal Information Dynamics in Spiking Neural Networks (44 citations)
- Complex Dynamic Neurons Improved Spiking Transformer Network for Efficient Automatic Speech Recognition (28 citations)
- Astromorphic Self-Repair of Neuromorphic Hardware Systems (13 citations)
- Deep Spiking Neural Networks with High Representation Similarity Model Visual Pathways of Macaque and Mouse (8 citations)
Quantization · 4 papers
- OMPQ: Orthogonal Mixed Precision Quantization (26 citations)
- Adaptive Low-Precision Training for Embeddings in Click-Through Rate Prediction (21 citations)
- Query-Aware Quantization for Maximum Inner Product Search (9 citations)
- Fast and Accurate Binary Neural Networks Based on Depth-Width Reshaping (3 citations)
Mixture of experts · 1 paper
- MoEC: Mixture of Expert Clusters (14 citations)