Adaptive Quantization

Follow this topic Ask about this topic

Latest papers

19 papers5 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Not All Frames Are Equal:Difference-Aware Quantization for Ultra-Low-Bit ASR
Oct 14, 2026Woori Jeon, Jungmin SoLow-Bit Quantization Automatic Speech Recognition
Sogang University, Seoul, Korea
TETRIS-4DGS: Efficient Post-Training Compression of 4D Gaussian Splatting
Oct 10, 2026Chengzhen Wu, Kelei Liu, Sicheng Li +2Gaussian Splatting Compression Adaptive Quantization
Zhejiang University
Q-PACE: Dynamic Precision Allocation for Quantization-Aware Training
Oct 6, 2026Alexandra Volkova, Matin Ansaripour, Erik Schultheis +2LLM Quantization Quantization-Aware Training
ISTA · EPFL, ISTA
Beyond Uniform Compression: Budgeted Transmission Allocation for Extreme Federated Learning
Sep 30, 2026Pengfei Li, Mohammad KhalilCommunication-Efficient Distributed Training Low-Rank Compression
Centre for the Science of Learning & Technology (SLATE), University of Bergen, Bergen, Norway
Precision As You Need: Stochastic Computing Is a Dense Adaptive Quantizer
Sep 26, 2026Haoran Jin, Kangqi Zhang, Jirong Yang +4Adaptive Quantization Efficient Transformer Inference
Department of Electrical Engineering and Computer Science, University of Michigan
Fathom: Per-Query Read Depth for Sparse Decoding over Offloaded KV Caches
Sep 15, 2026Vivek KalyanaranganKV-Cache Offloading KV Caching
AQLoRA: A Zero-Search Recipe for Fast Quantized LoRA Fine-Tuning
Aug 24, 2026Md Romyull IslamFine-Tuning Quantization-Aware LoRA
Kennesaw State University
Entropy-Constrained Adaptive Stochastic Quantization
Aug 10, 2026Ran Ben Basat, Yaniv Ben-Itzhak, Michael Mitzenmacher +1Adaptive Quantization Neural Network Quantization
University College London · VMware Research by Broadcom · Harvard University
APQF: Agentic Profiling-Guided Structured Pruning and Mixed-Precision Quantization with Adaptive Fine-Tuning
Aug 6, 2026Sadegh Jafari, Mohiuddin Bilwal, Fan Zhou +2Model Compression Mixed-Precision Quantization
Department of Computer Science Iowa State University Ames, IA, USA · Department of Agricultural and Biosystems Engineering Iowa State University Ames, IA, USA
TASQ: Temporal-Adaptive Bit Sparsification Quantization for Diffusion Models
Aug 4, 2026Seokho Han, Dongwei Wang, Jinhee Kim +4Weight-Only Quantization Mixed-Precision Quantization
Department of Electrical and Computer Engineering, Sungkyunkwan University, Korea · Department of Electrical and Computer Engineering, University of Arizona, USA · Department of Electrical and Computer Engineering, Duke University, USA +1
Differentiable Proxy Learning for Adaptive Quantization Control in H.264 Video Coding
Jul 11, 2026Qihan Xu, Ivan V. BajićLearned Image Compression Rate-Distortion Optimization
School of Engineering Science, Simon Fraser University Burnaby, BC, Canada
Variable Bit-width Quantization: Learning Per-Group Precision for "Bigger-but-Smaller" Language Models
Jul 3, 2026Hamish OgilvyLLM Inference Acceleration Mixed-Precision Quantization
Algolia Research
Inner Product Aware Quantization: Provably Fast, Accurate, and Adaptive Algorithms
May 29, 2026Nathan White, Krish SingalVector Quantization Neural Network Quantization
University of Pennsylvania
Training-Free Continuous Bitrate Control for Scalable Image Coding for Humans and Machines
May 29, 2026Yui Tatsumi, Hiroshi WatanabeLearned Image Compression Adaptive Quantization
Graduate School of FSE, Waseda University Tokyo, Japan
ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models
May 19, 2026Arash Akbari, Arman Akbari, Masih Eskandar +11Neural Network Quantization VLM Quantization
1Northeastern University · University of Georgia · 3Cisco Systems
Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy
Apr 25, 2026Emre Ardıç, Yakup GençNon-IID Federated Learning Differentially Private Federated Learning
Department of Computer Engineering, Gebze Technical University, Gebze, Kocaeli 41400, Türkiye
RateQuant: Optimal Mixed-Precision KV Cache Quantization via Rate-Distortion Theory
Apr 22, 2026Fei Zuo, Zikang Zhou, Hao Cong +2LLM Inference Efficiency Mixed-Precision Quantization
1BA TechWorks (BMW Group) · 2National University of Singapore · 3Tsinghua University
DeFakeQ: Enabling Real-Time Deepfake Detection on Edge Devices via Adaptive Bidirectional Quantization
Apr 10, 2026Xiangyu Li, Yujing Sun, Yuhang Zheng +2Adaptive Quantization Deepfake Detection
Digital Trust Centre, Nanyang Technological University · School of Information Science and Technology, ShanghaiTech University
You Had One Job: Per-Task Quantization Using LLMs' Hidden Representations
Nov 9, 2025Amit LeVi, Raz Lapid, Rom Himelstein +3Weight-Only Quantization Mixed-Precision Quantization
1Technion—Israel Institute of Technology · 2Intuit · 3Ben-Gurion University of the Negev +1

Oct 14, 2026J/K move · Enter open · S save

Not All Frames Are Equal:Difference-Aware Quantization for Ultra-Low-Bit ASR

Woori Jeon, Jungmin So

Sogang University, Seoul, Korea

Low-Bit Quantization Automatic Speech Recognition Post-Training Quantization Adaptive Quantization

Open paper AskPDF arXiv

Post-training quantization (PTQ) methods such as GPTQ and AWQ compress large language models effectively, but applying them to automatic speech recognition (ASR) at ultra-low bit-widths (2-3 bits) causes severe transcription degradation. Unlike discrete text tokens, speech activations are highly correlated in steady-state and zero-padded regions yet change rapidly at phonetic boundaries; when all frames contribute equally to the Hessian, static frames dominate calibration and obscure informative transitions. We propose DiffAQ, which computes frame-to-frame activation differences to measure the rate of acoustic change and assigns Hessian importance proportionally, concentrating quantization precision on phonetically critical frames. As a training-free modification to GPTQ, DiffAQ consistently reduces WER across various Whisper sizes and standard benchmarks, with the largest gains at 2-bit precision where baseline methods frequently produce degenerate outputs.