No article text was recorded for this version.
non-linear operation bottleneck
Version 1 of 146 papers61 conceptsfrozen 2026-10-04
Papers in this version
- Blog: SoftmAP: Software-Hardware Co-design for Integer-Only Softmax on Associative Processors
- Blog: MA-BERT: Towards Matrix Arithmetic-only BERT Inference by Eliminating Complex Non-Linear Functions
- Blog: Hyft: A Reconfigurable Softmax Accelerator with Hybrid Numeric Format for both Training and Inference
- Blog: QUARK: Quantization-Enabled Circuit Sharing for Transformer Acceleration by Exploiting Common Patterns in Nonlinear Operations
- Blog: SSR: Spatial Sequential Hybrid Architecture for Latency Throughput Tradeoff in Transformer Acceleration
- Blog: MOSS: Efficient and Accurate FP8 LLM Training with Microscaling and Automatic Scaling
- Blog: Softermax: Hardware/Software Co-Design of an Efficient Softmax for Transformers
- Blog: SwiftTron: An Efficient Hardware Accelerator for Quantized Transformers
- Blog: PackQViT: Faster Sub-8-bit Vision Transformers via Full and Packed Quantization on the Mobile
- Blog: PIVOT- Input-aware Path Selection for Energy-efficient ViT Inference
- Blog: TurboTransformers: An Efficient GPU Serving System For Transformer Models
- Blog: VEXP: A Low-Cost RISC-V ISA Extension for Accelerated Softmax Computation in Transformers
- Blog: Full Stack Optimization of Transformer Inference
- Blog: Towards Zero-Stall Matrix Multiplication on Energy-Efficient RISC-V Clusters for Machine Learning Acceleration
- Blog: LightSeq: A High Performance Inference Library for Transformers
- Blog: GSQ-Tuning: Group-Shared Exponents Integer in Fully Quantized Training for LLMs On-Device Fine-tuning
- Blog: ITERA-LLM: Boosting Sub-8-Bit Large Language Model Inference via Iterative Tensor Decomposition
- Blog: NOVA: NoC-based Vector Unit for Mapping Attention Layers on a CNN Accelerator
- Blog: SqueezeBERT: What can computer vision teach NLP about efficient neural networks?
- Blog: ERQ: Error Reduction for Post-Training Quantization of Vision Transformers
- Blog: APTQ: Attention-aware Post-Training Mixed-Precision Quantization for Large Language Models
- Blog: HAAN: A Holistic Approach for Accelerating Normalization Operations in Large Language Models
- Blog: NLI : Non-uniform Linear Interpolation Approximation of Nonlinear Operations for Efficient LLMs Inference
- Blog: Win Fast or Lose Slow: Balancing Speed and Accuracy in Latency-Sensitive Decisions of LLMs
- Blog: HiT-SR: Hierarchical Transformer for Efficient Image Super-Resolution
- Blog: DRONE: Data-aware Low-rank Compression for Large NLP Models
- Blog: BBAL: A Bidirectional Block Floating Point-Based Quantisation Accelerator for Large Language Models
- Blog: NN-LUT: Neural Approximation of Non-Linear Operations for Efficient Transformer Inference
- Blog: Pushing the Limits of BFP on Narrow Precision LLM Inference
- Blog: FQ-PETR: Fully Quantized Position Embedding Transformation for Multi-View 3D Object Detection
- Blog: Modeling the Resource Requirements of Convolutional Neural Networks on Mobile Devices
- Blog: Topkima-Former: Low-energy, Low-Latency Inference for Transformers using top-k In-memory ADC
- Blog: Genetic Quantization-Aware Approximation for Non-Linear Operations in Transformers
- Blog: EXAQ: Exponent Aware Quantization For LLMs Acceleration
- Blog: I-ViT: Integer-only Quantization for Efficient Vision Transformer Inference
- Blog: Linear Computation Coding
- Blog: WK, WV is (Linearly) All You Need: On the Necessity of the QKV Weight Triplet in Self-Attention Transformers
- Blog: Learning to Quantize Deep Networks by Optimizing Quantization Intervals With Task Loss
- Blog: Compressing Large-Scale Transformer-Based Models: A Case Study on BERT
- Blog: Auto-ViT-Acc: An FPGA-Aware Automatic Acceleration Framework for Vision Transformer with Mixed-Scheme Quantization
- Blog: TATAA: Programmable Mixed-Precision Transformer Acceleration with a Transformable Arithmetic Architecture
- Blog: Hardware Acceleration of Fully Quantized BERT for Efficient Natural Language Processing
- Blog: HyC-LoRA: Memory Efficient LoRA Fine-tuning with Hybrid Activation Compression
- Blog: When to Use Efficient Self Attention? Profiling Text, Speech and Image Transformer Variants
- Blog: ITA: An Energy-Efficient Attention and Softmax Accelerator for Quantized Transformers
- Blog: SOLE: Hardware-Software Co-design of Softmax and LayerNorm for Efficient Transformer Inference
History & citation
Current · Permanent link
Cite this direction
Latest version (a living page):
@misc{lacuna_non_linear_operation_bottleneck_590559f6,
title = {non-linear operation bottleneck},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/non-linear-operation-bottleneck/rdr_590559f668674577a8ace430139a39a6},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}This exact version (version 1, frozen text):
@misc{lacuna_non_linear_operation_bottleneck_590559f6_v1,
title = {non-linear operation bottleneck},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/non-linear-operation-bottleneck/rdr_590559f668674577a8ace430139a39a6/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}