AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Two-dimensional quantization for geometry-aware audio coding

Tal Shuster, Eliya Nachmani

ICML 2026 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use rhombic grid instead of rectangle grid in Q2D2 (ablation)Q2D2 · LibriTTS test-cleanPESQ: 2.2909 -> 2.3995Improved
use rhombic grid instead of hexagonal grid in Q2D2 (ablation)Q2D2 · LibriTTS test-cleanPESQ: 2.2862 -> 2.3995Improved
use 6 latent dimensions instead of 8 (lower resolution) at 1 kbps (ablation)Q2D2 · LibriTTS test-cleanPESQ: 2.092 -> 2.3995Improved
use 6 latent dimensions instead of 4 (higher resolution) at 1 kbps (ablation)Q2D2 · LibriTTS test-cleanPESQ: 2.018 -> 2.3995Improved
increase grid levels from [7]x6 to [11]x6 (ablation)Q2D2 · LibriTTS test-cleanPESQ: 2.3995 -> 4.2832 PESQ; codebook utilization 92.18% -> 72.11%Mixed
replace VQ (WavTokenizer) quantizer with Q2D2 in same framework (ablation)WavTokenizer framework · LibriTTS test-clean (LibriTTS-trained)PESQ: 2.4655 -> 2.5021Improved
replace FSQ with Q2D2 in same framework (ablation)WavTokenizer framework · LibriTTS test-clean (LibriTTS-trained)PESQ: 2.3873 -> 2.5021Improved
Q2D2 at 1 kbps vs WavTokenizer on LJSpeech (out-of-domain)Q2D2 1kbps · LJSpeechUTMOS: 4.2580 -> 3.9715Hurt
use Q2D2 single quantizer (3.3 kbps) instead of multi-quantizer DAC 9 kbpscodec · MUSHRAMUSHRA score: 92.64 -> 98.05Improved
use Q2D2 tokens instead of WavTokenizer tokens for TTSParlerTTS 600M · LibriTTS TTSCMOS-Q: -0.22 -> -0.11Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy