Different Tasks, Different Timesteps: Temporal Task-Aware Quantization for Diffusion Language Models

Ofek Avivi, Eliron Rahimi, Avi Mendelson, Chaim Baskin

Different Tasks, Different Timesteps: Temporal Task-Aware Quantization for Diffusion Language Models: 21 upvotes on Hugging Face Daily Papers, #32 of 50 papers on 2026-10-23. Day-by-day upvote history.

Diffusion Language Models (DLMs) introduce unique quantization challenges due to their iterative denoising process. In this work, we reveal that quantization sensitivity is both taskand time-dependent, with different downstream tasks exhibiting distinct precision requirements across denoising steps. We further show that these patterns can be predicted without groundtruth labels or task-specific evaluation metrics. Building on these findings, we introduce T 2Quant, a task and time-aware mixedprecision quantization framework that uses a label-free sensitivity signal to optimally allocate activation precision across denoising steps under a fixed bit-width budget. Extensive experiments across three DLMs and six benchmarks demonstrate that T 2Quant consistently improves the average accuracy of existing quantization methods, achieving gains of up to 6.7 percentage points under the same budget.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.