paper-with-me

Papers

SAID: Accelerating Diffusion-Based Language Models via Scaffold-Aware Iterative Decoding

2026-06-03 · Na Li, Chengda Wang, Mingju Gao, Hao Tang arxiv

Diffusion large language models (DLLMs) enable non-autoregressive generation by iteratively denoising corrupted token sequences with bidirectional context. Despite their ability to update multiple positions in parallel, inference remains costly due to the many denoising steps required for high-quality generation. We propose SAID, a Scaffold-Aware Iterative Decoding framework that accelerates DLLMs by reallocating computation across tokens. SAID first spends denoising computation on scaffold tokens to establish the coarse semantic structure, and then completes predictable detail tokens with fewer steps. We further adapt SAID to block-wise diffusion decoding and introduce Confidence-Hierarchical Layered Generation (CHLG), which assigns additional steps only to low-confidence tokens. Experiments on LLaDA-8B and LLaDA 1.5 across math, coding, and knowledge benchmarks show that SAID significantly accelerates DLLM inference with a maximum speedup of 9.1x while maintaining competitive performance. Our code is publicly available: https://github.com/TH-AI-Lab-PKU/SAID.

📄 PDF Abstract BibTeX arXiv:2606.04974

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion

2025-08-26 · Justin Jung arxiv

Generating realistic sparse multi-category 3D voxel structures is difficult due to the cubic memory scaling of voxel structures and moreover the significant class imbalance caused by sparsity. We introduce Scaffold Diffu…

Autoregressive fragment-based diffusion for pocket-aware ligand design

2023-12-15 · Mahdi Ghorbani, Leo Gendelev, Paul Beroza, Michael J. Keiser

In this work, we introduce AutoFragDiff, a fragment-based autoregressive diffusion model for generating 3D molecular structures conditioned on target protein structures. We employ geometric vector perceptrons to predict …

Scaffold-Aware Generative Augmentation and Reranking for Enhanced Virtual Screening

2025-10-18 · Xin Wang, Yu Wang, Yunchao Liu, Jens Meiler 외 arxiv

Ligand-based virtual screening (VS) is an essential step in drug discovery that evaluates large chemical libraries to identify compounds that potentially bind to a therapeutic target. However, VS faces three major challe…

Drug Discovery

Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problem

2022-06-08 · Brian L. Trippe, Jason Yim, Doug Tischer, David Baker 외

Construction of a scaffold structure that supports a desired motif, conferring protein function, shows promise for the design of vaccines and enzymes. But a general solution to this motif-scaffolding problem remains open…

Graph Neural Network

DiffHopp: A Graph Diffusion Model for Novel Drug Design via Scaffold Hopping

2023-08-14 · Jos Torge, Charles Harris, Simon V. Mathis, Pietro Lio

Scaffold hopping is a drug discovery strategy to generate new chemical entities by modifying the core structure, the \emph{scaffold}, of a known active compound. This approach preserves the essential molecular features o…

Drug DesignDrug Discovery