From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs

• URL: http://arxiv.org/abs/2512.06776v1
• Date: Sun, 07 Dec 2025 10:28:21 GMT
• Title: From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs
• Abstract summary: We show that principled AR-to-block-diffusion adaptation is an effective and compute-efficient alternative to training DLMs from scratch.<n> NBDiff-7B (Base and Instruct) could inherit the long-context modeling and reasoning capabilities, and achieve state-of-the-art performance.
• Score: 58.640039233470766
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) excel at generation but dominant autoregressive (AR) decoding is inherently sequential, creating a throughput bottleneck. Diffusion Language Models (DLMs)--especially block-wise variants--enable parallel generation and intra-block bidirectional reasoning, yet training large DLMs from scratch is costly and wastes the knowledge in mature AR checkpoints. Prior "adaptation" attempts either modify logits or randomly grow attention masks to full-sequence diffusion, or simply transplant AR weights into a block-diffusion recipe, leaving a fundamental mismatch between AR causality and block-wise bidirectionality unaddressed. We reframe adaptation as a intra-paradigm path from AR to Block-Diffusion by viewing AR as Block-Diffusion with blocksize=1. Concretely, we design the pathway of adaptation as follows: we use a context-causal attention mask (causal in context, bidirectional only within the active block), an efficient parallel adaptation procedure, an auxiliary AR loss to maximize data utilization and retain pretrained knowledge, and gradual increment of the generation block size. The recipe integrates cleanly with masked block-diffusion and maintains train-inference consistency. Built on these components, NBDiff-7B (Base and Instruct) could inherit the long-context modeling and reasoning capabilities, and achieve state-of-the-art performance among the 7B-class DLMs, delivering strong gains on general-knowledge, math, and code benchmarks over strong baselines. These results demonstrate that principled AR-to-block-diffusion adaptation is an effective and compute-efficient alternative to training DLMs from scratch. Codes: https://github.com/YuchuanTian/NBDiff.

Related papers

• FLARE: Diffusion for Hybrid Language Model [72.60770374799634]
  FLARE is a systematic conversion framework for hybrid-attention large language models.<n>It combines a token-equal AR-and-diffusion objective, hardware-aware kernels, and unified inference, enabling one checkpoint to support both AR-style verified decoding and diffusion-style parallel denoising.<n>Our results suggest practical dLLMs are limited not only by decoding algorithms, but also by transfer data quality and the training inefficiency of current block-diffusion objectives.
  arXiv  Detail & Related papers  (2026-06-01T06:58:15Z)
• BlockVLA: Accelerating Autoregressive VLA via Block Diffusion Finetuning [41.5997751218601]
  BlockVLA is a framework that adapts pretrained AR backbones into an efficient discrete diffusion policy.<n>We conduct extensive evaluations on the LIBERO and SimplerEnv benchmarks.<n>Our model exhibits superior training efficiency, with success rates converging substantially faster than baselines.
  arXiv  Detail & Related papers  (2026-05-13T11:37:51Z)
• From AR to Diffusion: Efficiently Adapting Large Language Models with Strictly Causal and Elastic Horizons [74.25767873168745]
  Diffusion models promise efficient parallel text generation but rely on bidirectional attention, creating a structural mismatch with pre-trained Autoregressive (AR) models.<n>We propose FLUID, a framework that efficiently adapts AR backbones to the diffusion paradigm.<n> Experiments demonstrate that FLUID achieves state-of-the-art performance while reducing training costs by orders of magnitude.
  arXiv  Detail & Related papers  (2026-04-11T13:18:27Z)
• Causal Autoregressive Diffusion Language Model [70.7353007255797]
  CARD reformulates the diffusion process within a strictly causal attention mask, enabling dense, per-token supervision in a single forward pass.<n>Our results demonstrate that CARD achieves ARM-level data efficiency while unlocking the latency benefits of parallel generation.
  arXiv  Detail & Related papers  (2026-01-29T17:38:29Z)
• Efficient-DLM: From Autoregressive to Diffusion Language Models, and Beyond in Speed [76.49335677120031]
  Diffusion language models (dLMs) have emerged as a promising paradigm that enables parallel, non-autoregressive generation.<n>We study AR-to-dLM conversion to transform pretrained AR models into efficient dLMs that excel in speed while preserving AR models' task accuracy.
  arXiv  Detail & Related papers  (2025-12-16T04:12:17Z)
• SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generation [62.14510717860079]
  We propose a Synergistic Diffusion-Autoregression paradigm that unifies the training efficiency of autoregressive models with the parallel inference capability of diffusion.<n>SDAR performs a lightweight paradigm conversion that transforms a well-trained autoregressive (AR) model into a blockwise diffusion model through brief, data-efficient adaptation.<n>Building on this insight, SDAR achieves efficient AR-to-diffusion conversion with minimal cost, preserving AR-level performance while enabling parallel generation.
  arXiv  Detail & Related papers  (2025-10-07T17:29:28Z)
• AdaBlock-dLLM: Semantic-Aware Diffusion LLM Inference via Adaptive Block Size [7.442463267121892]
  Diffusion-based large language models (dLLMs) are gaining attention for their inherent capacity for parallel decoding.<n>This paper presents the first systematic investigation challenging the fixed block size assumption in semi-AR decoding.<n>We introduce AdaBlock-dLLM, a training-free, plug-and-play scheduler that adaptively aligns block boundaries with semantic steps by adjusting block size during runtime.
  arXiv  Detail & Related papers  (2025-09-30T15:53:56Z)
• Fast-dLLM v2: Efficient Block-Diffusion LLM [64.38006546510337]
  Fast-dLLM v2 is a block diffusion language model that adapts pretrained AR models into dLLMs for parallel text generation.<n>This represents a 500x reduction in training data compared to full-attention diffusion LLMs such as Dream (580B tokens)
  arXiv  Detail & Related papers  (2025-09-30T14:40:18Z)
• Taming Masked Diffusion Language Models via Consistency Trajectory Reinforcement Learning with Fewer Decoding Step [28.12392773921128]
  Masked diffusion language models offer properties such as parallel decoding, flexible generation orders, and the potential for fewer inference steps.<n>A naive approach is to directly transfer techniques well-established for autoregressive (AR) language models to MDLMs.<n>We propose EOS Early Rejection (EOSER) and Ascending Step-Size (ASS) decoding schedulers, which unlock the potential of MDLMs to perform full diffusion-style decoding.
  arXiv  Detail & Related papers  (2025-09-28T15:01:15Z)
• Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive Decoding [60.06816407728172]
  Discrete diffusion language models have shown strong potential for text generation.<n>Standard supervised fine-tuning misaligns with semi-autoregressive inference.<n>We propose Blockwise SFT, which partitions responses into fixed-size blocks.
  arXiv  Detail & Related papers  (2025-08-27T02:49:33Z)
• DiffusionBlocks: Block-wise Neural Network Training via Diffusion Interpretation [11.910667302899638]
  DiffusionBlocks is a principled framework for transforming transformer-based networks into genuinely independent trainable blocks.<n>Our experiments on a range of transformer architectures demonstrate that DiffusionBlocks training matches the performance of end-to-end training.
  arXiv  Detail & Related papers  (2025-06-17T05:44:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.