Diffusion Language Model Inference

Overview

Diffusion language model inference studies how masked or continuous text representations can be refined over multiple steps, often updating many tokens in parallel rather than generating strictly left to right. Common approaches adapt which tokens to reveal or commit, reuse cached attention and activations, prune redundant computation, and use speculative decoding or verification to reduce denoising steps. Evaluation spans reasoning and code benchmarks, long-context tasks, multimodal generation, and serving measures such as latency, throughput, and energy use. The central tension is that more parallelism and caching can save computation but may harm coherence or accuracy, so current work seeks adaptive methods that balance speed with reliable generation across tasks and hardware.

Latest papers

201 papers31 papers in the last four weeks, down 6% on the four weeks before. 0.2% of all new papers.

Feb 23, 2026

Adaptation to Intrinsic Dependence in Diffusion Language Models

Diffusion language models (DLMs) have recently emerged as a promising alternative to autoregressive (AR) approaches, enabling parallel token generation beyond a rigid left-to-right order. Despite growing empirical success, the theoretical understanding of how unmasking schedules -- which specify the order and size of unmasked tokens during sampling -- affect generation quality remains limited. In this work, we introduce a distribution-agnostic unmasking schedule for DLMs that adapts to the (unknown) dependence structure of the target data distribution, without requiring any prior knowledge or hyperparameter tuning. In contrast to prior deterministic procedures that fix unmasking sizes, our method randomizes the number of tokens revealed at each iteration. We show that, for two specific parameter choices, the sampling convergence guarantees -- measured by Kullback-Leibler (KL) divergence -- scale as O~(TC/K)\widetilde O(\mathsf{TC}/K) and O~(DTC/K)\widetilde O(\mathsf{DTC}/K) respectively. Here, KK is the number of iterations, and TC\mathsf{TC} and DTC\mathsf{DTC} are the total correlation and dual total correlation of the target distribution, capturing the intrinsic dependence structure underlying the data. Importantly, our guarantees hold in the practically relevant parallel-sampling regime K<LK<L where LL is the token sequence length. These results significantly improve upon prior convergence theories and yield substantial sampling acceleration for low-complexity distributions. Overall, our findings unveil the adaptivity of DLMs to intrinsic data structures and shed light on the benefit of randomized unmasking sizes in inference schedule design.
Feb 22, 2026

IDLM: Inverse-distilled Diffusion Language Models

Diffusion Language Models (DLMs) have recently achieved strong results in text generation. However, their multi-step sampling leads to slow inference, limiting practical use. To address this, we extend Inverse Distillation, a technique originally developed to accelerate continuous diffusion models, to the discrete setting. Nonetheless, this extension introduces both theoretical and practical challenges. From a theoretical perspective, the inverse distillation objective lacks uniqueness guarantees, which may lead to suboptimal solutions. From a practical standpoint, backpropagation in the discrete space is non-trivial and often unstable. To overcome these challenges, we first provide a theoretical result demonstrating that our inverse formulation admits a unique solution, thereby ensuring valid optimization. We then introduce gradient-stable relaxations to support effective training. As a result, experiments on multiple DLMs show that our method, Inverse-distilled Diffusion Language Models (IDLM), reduces the number of inference steps by 4x-64x, while preserving the teacher model's generation quality. We provide the code, model checkpoints, and video tutorials on the project page: https://david-cripto.github.io/idlm-project-page
Feb 20, 2026

Evo: Autoregressive-Diffusion Large Language Models with Evolving Balance

We introduce Evo, a duality latent trajectory model that bridges autoregressive (AR) and diffusion-based language generation within a continuous evolutionary generative framework. Rather than treating AR decoding and diffusion generation as separate paradigms, Evo reconceptualizes text generation as a latent flow: each token is associated with a vector-valued embedding that evolves over a progression variable ti∈[0,1]t_i \in [0, 1], indicating its semantic maturity. Low tit_i values correspond to confident AR-like refinement, while high values invoke diffusion-style planning, allowing the model to adaptively balance AR and diffusion based on uncertainty. Theoretically, we show that both AR and diffusion models emerge as discretizations of a shared probability flow, and we derive Evo's training objective from a unified variational ELBO. The model is implemented as a time-conditioned Transformer governed by a shared vector field, trained end-to-end to jointly infer latent codes and their progression times. During decoding, Evo performs efficient, semantics-aware refinement, achieving high-quality outputs without sacrificing speed. Empirically, Evo 8B achieves state-of-the-art or highly competitive results on 15 diverse benchmarks, including reasoning (GSM8K, ARC-C), code generation (HumanEval, MBPP), and general language understanding, while maintaining fast inference speed. Our results demonstrate that Evo delivers a new paradigm for LLM design with strong generation quality, robust symbolic reasoning, and decoding efficiency.
Feb 19, 2026

Sink-Aware Pruning for Diffusion Language Models

Diffusion Language Models (DLMs) incur high inference cost due to iterative denoising, motivating efficient pruning. Existing pruning heuristics largely inherited from autoregressive (AR) LLMs, typically preserve attention sink tokens because AR sinks serve as stable global anchors. We show that this assumption does not hold for DLMs: the attention-sink position exhibits substantially higher variance over the full generation trajectory (measured by how the dominant sink locations shift across timesteps), indicating that sinks are often transient and less structurally essential than in AR models. Based on this observation, we propose Sink-Aware Pruning{\bf \texttt{Sink-Aware Pruning}}, which automatically identifies and prunes unstable sinks in DLMs (prior studies usually keep sinks for AR LLMs). Without retraining, our method achieves a better quality-efficiency trade-off and outperforms strong prior pruning baselines under matched compute. Our code is available at https://github.com/VILA-Lab/Sink-Aware-Pruning.
Feb 16, 2026

Scaling Beyond Masked Diffusion Language Models

Diffusion language models are a promising alternative to autoregressive models due to their potential for faster generation. Among discrete diffusion approaches, Masked diffusion currently dominates, largely driven by strong perplexity on language modeling benchmarks. In this work, we present the first scaling law study of uniform-state and interpolating discrete diffusion methods. We also show that Masked diffusion models can be made approximately 12% more FLOPs-efficient when trained with a simple cross-entropy objective. We find that perplexity is informative within a diffusion family but can be misleading across families, where models with worse likelihood scaling may be preferable due to faster and more practical sampling, as reflected by the speed-quality Pareto frontier. These results challenge the view that Masked diffusion is categorically the future of diffusion language modeling and that perplexity alone suffices for cross-algorithm comparison. Scaling all methods to 1.7B parameters, we show that uniform-state diffusion remains competitive on likelihood-based benchmarks and outperforms autoregressive and Masked diffusion models on GSM8K, despite worse validation perplexity. We provide the code, model checkpoints, and video tutorials on the project page: http://s-sahoo.github.io/scaling-dllms
Feb 11, 2026

Just on Time: Token-Level Early Stopping for Diffusion Language Models

Diffusion language models generate text through iterative refinement, a process that is often computationally inefficient because many tokens reach stability long before the final denoising step. We introduce a training-free, token-level early stopping approach that identifies convergence independently at each position. Our method leverages lightweight signals derived from the model's predictions and local context to dynamically determine when individual tokens can be finalized. This yields adaptive per-token freezing without task-specific fine-tuning, substantially reducing the total number of diffusion steps required. Across diverse benchmarks, spanning mathematical reasoning, general question answering, and scientific understanding, our approach achieves substantial efficiency gains while preserving generation quality.
Feb 6, 2026

DAWN: Dependency-Aware Fast Inference for Diffusion LLMs

Diffusion large language models (dLLMs) have shown advantages in text generation, particularly due to their inherent ability for parallel decoding. However, constrained by the quality--speed trade-off, existing inference solutions adopt conservative parallel strategies, leaving substantial efficiency potential underexplored. A core challenge is that parallel decoding assumes each position can be filled independently, but tokens are often semantically coupled. Thus, the correct choice at one position constrains valid choices at others. Without modeling these positional dependencies, parallel strategies produce deteriorated outputs. Motivated by this insight, we propose DAWN, a training-free, dependency-aware decoding method for fast dLLM inference. DAWN extracts dependency relations and leverages two key observations: (1) masked positions dependent on unmasked certain positions become more reliable; (2) coupled masked positions strongly influence each other's predictions. Given those findings, DAWN leverages a dependency graph to select more reliable unmasking positions at each iteration, achieving high parallelism with negligible loss in generation quality. Extensive experiments across multiple models and datasets demonstrate that DAWN speedups the inference by 1.80-8.06x over baselines while preserving the generation quality. Code is released at https://github.com/lizhuo-luo/DAWN.
Feb 2, 2026

Self-Rewarding Sequential Monte Carlo for Masked Diffusion Language Models

This work presents self-rewarding sequential Monte Carlo (SMC), an inference-time scaling algorithm enabling effective sampling of masked diffusion language models (MDLMs). Our algorithm stems from the observation that most existing MDLMs rely on a confidence-based sampling strategy, where only tokens with the highest prediction confidence are preserved at each step. This restricts the generation to a noise-sensitive, greedy decoding paradigm, resulting in an inevitable collapse in the diversity of possible paths. We address this problem by launching multiple interacting diffusion processes in parallel, referred to as particles, for trajectory exploration. Importantly, we introduce the trajectory-level confidence as a self-rewarding signal for assigning particle importance weights. During sampling, particles are iteratively weighted and resampled to systematically steer generation towards globally confident, high-quality samples. Our self-rewarding SMC is verified on various masked diffusion language models and benchmarks, achieving significant improvement without extra training or reward guidance, while effectively converting parallel inference capacity into improved sampling quality. Our code is available at https://github.com/Algolzw/self-rewarding-smc.
Jan 29, 2026

ILRR: Inference-Time Steering Method for Masked Diffusion Language Models

Discrete Diffusion Language Models (DLMs) offer a promising non-autoregressive alternative for text generation, yet effective mechanisms for inference-time control remain relatively underexplored. Existing approaches include sampling-level guidance or trajectory optimization mechanisms. In this work, we study the paradigm of reference-based latent steering for DLMs. We introduce Iterative Latent Representation Refinement (ILRR), an efficient framework for steering DLMs using a reference text as a high-level semantic blueprint. ILRR extracts and injects reference-derived semantic signals into the evolving activations of the generated sequence, enabling tunable transfer of coarse properties such as sentiment. We further introduce Spatially Modulated Steering, an extension that enables long-form generation to be guided by shorter references by regulating intensity across the sequence. Empirically, we demonstrate that ILRR achieves effective control on LLaDA and MDLM architectures with low computational overhead, requiring only one additional parallel forward pass per denoising step. Under comparable compute budgets, ILRR improves attribute accuracy over baselines by 10% to 60% points. Our results suggest that the iterative, global denoising process makes DLMs a natural substrate for effective sequence-wide activation-level control.
Jan 28, 2026

Window-Diffusion: Accelerating Diffusion Language Model Inference with Windowed Token Pruning and Caching

Diffusion language models (DLMs) generate text through iterative denoising, but inference requires full-sequence attention at every iteration, resulting in substantial redundant computation on masked tokens. Block-wise diffusion can reduce this cost, yet it typically relies on retraining and constrained update orders, limiting its direct applicability to pretrained DLMs. Our token-level analysis reveals pronounced structural locality in DLM inference. Decoding is driven by a small set of prefix-localized active tokens; the influence of distant undecoded context diminishes rapidly, and decoded tokens exhibit stage-wise temporal stability, enabling reuse of intermediate representations except for a brief post-decode transient. Motivated by these observations, we propose \placeholder\footnote{The source code is available at https://github.com/vhicrgit/Window-Diffusion.}, a window-based token pruning and caching method for inference. We maintain a local computation window that slides rightward as denoising progresses, and partition undecoded tokens into: (i) active tokens that are computed online, (ii) buffer tokens whose KV states are cached and periodically refreshed, and (iii) far-field tokens that are pruned outside the window. Computation is restricted to active and buffer tokens within the window, while far-field tokens are omitted at each stage. Experiments on LLaDA and Dream show that, under matched compute budgets, our method achieves up to 99×99\times inference speedup while largely preserving generation performance.
Jan 25, 2026

Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding

Diffusion Large Language Models (dLLMs) offer a compelling paradigm for natural language generation, leveraging parallel decoding and bidirectional attention to achieve superior global coherence compared to autoregressive models. While recent works have accelerated inference via KV cache reuse or heuristic decoding, they overlook the intrinsic inefficiencies within the block-wise diffusion process. Specifically, they suffer from spatial redundancy by modeling informative-sparse suffix regions uniformly and temporal inefficiency by applying fixed denoising schedules across all the decoding process. To address this, we propose Streaming-dLLM, a training-free framework that streamlines inference across both spatial and temporal dimensions. Spatially, we introduce attenuation guided suffix modeling to approximate the full context by pruning redundant mask tokens. Temporally, we employ a dynamic confidence aware strategy with an early exit mechanism, allowing the model to skip unnecessary iterations for converged tokens. Extensive experiments show that Streaming-dLLM achieves up to 68.2X speedup while maintaining generation quality, highlighting its effectiveness in diffusion decoding. The code is available at https://github.com/xiaoshideta/Streaming-dLLM.
Jan 10, 2026

Mosaic: Unlocking Over 30×\times Context Length for Diffusion LLMs Inference via Global Memory Planning and Dynamic Peak Taming

Diffusion-based large language models (dLLMs) have emerged as a promising alternative to autoregressive models, leveraging simultaneous denoising to enable global planning and iterative refinement. These properties make dLLMs attractive for long-context generation. However, deploying dLLMs faces a prohibitive memory barrier, as existing inference systems are inefficient for the diffusion paradigm. We observe that current inference systems are misaligned with dLLMs. Unlike autoregressive models, whose memory footprint is dominated by the KV-Cache, dLLMs are bottlenecked by transient activations rematerialized per step. Moreover, generic memory reuse mechanisms lack the global visibility to handle dynamic memory peaks of dLLMs, which alternate between logits and feed-forward networks. To address these challenges, we present Mosaic, a memory-efficient inference system that shifts dLLM execution from local, static management to a global, dynamic paradigm. Mosaic integrates (i) a mask-only logits kernel eliminating redundant activation materialization, (ii) a lazy chunking optimizer using online heuristics to tame dynamic memory peaks, and (iii) a global memory manager leveraging virtual addressing to mitigate memory fragmentation. Evaluations show that Mosaic reduces the memory peak-to-average ratio by 2.71×\times on average and increases the maximum inference sequence length on identical hardware by 15.30-32.34×\times. Crucially, Mosaic is training-free and preserves exact model outputs, while reducing end-to-end latency by 2.5%-55.4%. Our code is publicly available at https://github.com/flashserve/Mosaic.
Jan 6, 2026

Systematic Hazard Sampling: Minimal-Variance Inference for Discrete Diffusion and Flow Models

Uniform-noise discrete diffusion and flow models generate sequences non-autoregressively through iterative, context-dependent token replacements. However, these models are typically formulated as time-inhomogeneous continuous- or discrete-time Markov chains (CTMC/DTMC), sampled using independent Bernoulli change decisions per discretization step. This induces Poisson-binomial variance in per-position jump counts that grows with the number of required edits, leading to the common under-editing (residual noise) and over-editing (cascading substitutions) failure modes that degrade sample quality. We identify this sampler-induced variance as an orthogonal source of degradation, distinct from model-side errors and addressable purely at inference time. We propose Systematic Hazard Sampling (SHS), a training-free, drop-in, and hyperparameter-free inference principle for any sampler that admits a stay-vs.-replace decomposition. SHS models per-token edits as events driven by cumulative hazard (CTMC) or jump mass (DTMC) and triggers an edit whenever this quantity exceeds unit-spaced thresholds with a single random phase per position. For any fixed cumulative mass, this preserves the expected jump count while achieving the minimum conditional variance possible among unbiased integer estimators (at most 1/4), without altering per-jump destination sampling. Experiments on four uniform-noise discrete diffusion and flow language models spanning ~110M to ~3B parameters show that SHS consistently improves sample quality across numbers of function evaluations (NFE). Code is available at https://github.com/Jang-seunghwan/Systematic-Hazard-Sampling.
Jan 5, 2026

CD4LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models

Diffusion language models (DLMs) accelerate generation by decoding multiple tokens in parallel, but pushing them to fast decoding often sacrifices accuracy. We attribute this to a static-to-dynamic mismatch: training optimizes local denoising under fixed schedules, while fast inference skips steps and encounters irregular mask patterns. Existing accelerators largely address only one side of this problem. Training-free inference methods speed up decoding without changing the model, leaving the mismatch unresolved; training-side distillation methods adapt the model, but often tie the student to a specific decoding trajectory, limiting composability and still suffering accuracy loss. We propose CD4LM, a composable framework for accurate DLM acceleration. Its core training component, Discrete-Space Consistency Distillation (DSCD) trains a trajectory-invariant student on diverse masked states, reducing the mismatch without fixing a decoding policy. We pair DSCD with a lightweight ConfidenceAdaptive Decoder, while keeping the trained student compatible with external inference-side accelerators. On GSM8K, CD4LM preserves accuracy while achieving a 5.18x speedup; stacking CD4LM with per-step cache accelerators further pushes speedup to 9.73x at teacher-level accuracy, suggesting a scalable path for combining training-side robustness with future decoding advances. Code is available at https://github.com/yihao-liang/CDLM.
Dec 16, 2025

Sparse-LaViDa: Sparse Multimodal Discrete Diffusion Language Models

Masked Discrete Diffusion Models (MDMs) have achieved strong performance across a wide range of multimodal tasks, including image understanding, generation, and editing. However, their inference speed remains suboptimal due to the need to repeatedly process redundant masked tokens at every sampling step. In this work, we propose Sparse-LaViDa, a novel modeling framework that dynamically truncates unnecessary masked tokens at each inference step to accelerate MDM sampling. To preserve generation quality, we introduce specialized register tokens that serve as compact representations for the truncated tokens. Furthermore, to ensure consistency between training and inference, we design a specialized attention mask that faithfully matches the truncated sampling procedure during training. Built upon the state-of-the-art unified MDM LaViDa-O, Sparse-LaViDa achieves up to a 2x speedup across diverse tasks including text-to-image generation, image editing, and mathematical reasoning, while maintaining generation quality.
Nov 19, 2025

A Comprehensive Study on Visual Token Redundancy for Discrete Diffusion-based Multimodal Large Language Models

Discrete diffusion-based multimodal large language models (dMLLMs) have emerged as a promising alternative to autoregressive MLLMs thanks to their advantages in parallel decoding and bidirectional context modeling, but most existing dMLLMs incur significant computational overhead during inference due to the full-sequence attention computation in each denoising step. Pioneering studies attempt to resolve this issue from a modality-agnostic perspective via key-value cache optimization or efficient sampling but most of them overlook modality-specific visual token redundancy. In this work, we conduct a comprehensive study on how visual token redundancy evolves with different dMLLM architectures and tasks and how visual token pruning affects dMLLM responses and efficiency. Specifically, our study reveals that visual redundancy emerges only in from-scratch dMLLMs while handling long-answer tasks. In addition, we validate that visual token pruning introduces non-negligible information loss in dMLLMs and only from-scratch dMLLMs can recover the lost information progressively during late denoising steps. Furthermore, our study shows that layer-skipping is promising for accelerating AR-to-diffusion dMLLMs, whereas progressive or late-step pruning is more effective for from-scratch dMLLMs. Overall, this work offers a new perspective on efficiency optimization for dMLLMs, greatly advancing their applicability across various multimodal understanding tasks. The code is available at: https://github.com/Yrdal3910/dMLLM-Visual-Token-Redundancy-Analysis.
Oct 13, 2025

Latent Refinement Decoding: Enhancing Diffusion-Based Language Models by Refining Belief States

Autoregressive (AR) models remain the dominant paradigm for natural language generation, but their strictly sequential decoding process leads to high inference latency. Recent diffusion-inspired language models, such as LLaDA and Dream, alleviate this issue through parallel generation, yet they still face two key limitations: information loss, since predictive distributions over non-finalised tokens are discarded at each step, and unstable commitment dynamics, where local token decisions are not sufficiently coordinated at the global level. We propose Latent Refinement Decoding (LRD), a two-stage decoding framework consisting of Latent Refinement and a Predictive Feedback Loop. In the first stage, LRD keeps masked positions as distributional mixtures of predicted tokens and the mask embedding, enabling the model to form more globally consistent beliefs before committing tokens. In the second stage, LRD progressively finalises confident tokens while preserving uncertain ones for further iterative refinement, with KL-divergence dynamics serving as a stable criterion for convergence and early stopping. Experiments show that LRD improves performance on both coding tasks, including HumanEval by 6.3 and MBPP by 2.6, and reasoning tasks, including GSM8K by 2.9 and MATH500 by 3.8, while achieving up to 10.6x overall speedup. Moreover, LRD is compatible with single-token decoding, multi-token threshold commitment, and system-level accelerators; when combined with Fast-dLLM, it reaches up to 18.4x speedup over vanilla decoding while further improving accuracy.
Oct 6, 2025

ParallelBench: Understanding the Trade-offs of Parallel Decoding in Diffusion LLMs

While most autoregressive LLMs are constrained to one-by-one decoding, diffusion LLMs (dLLMs) have attracted growing interest for their potential to dramatically accelerate inference through parallel decoding. Despite this promise, the conditional independence assumption in dLLMs causes parallel decoding to ignore token dependencies, inevitably degrading generation quality when these dependencies are strong. However, existing works largely overlook these inherent challenges, and evaluations on standard benchmarks (e.g., math and coding) are not sufficient to capture the quality degradation caused by parallel decoding. To address this gap, we first provide an information-theoretic analysis of parallel decoding. We then conduct case studies on analytically tractable synthetic list operations from both data distribution and decoding strategy perspectives, offering quantitative insights that highlight the fundamental limitations of parallel decoding. Building on these insights, we propose ParallelBench, the first benchmark specifically designed for dLLMs, featuring realistic tasks that are trivial for humans and autoregressive LLMs yet exceptionally challenging for dLLMs under parallel decoding. Using ParallelBench, we systematically analyze both dLLMs and autoregressive LLMs, revealing that: (i) dLLMs under parallel decoding can suffer dramatic quality degradation in real-world scenarios, and (ii) current parallel decoding strategies struggle to adapt their degree of parallelism based on task difficulty, thus failing to achieve meaningful speedup without compromising quality. Our findings underscore the pressing need for innovative decoding methods that can overcome the current speed-quality trade-off. We release our benchmark to help accelerate the development of truly efficient dLLMs.
Sep 30, 2025

Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models

Diffusion Large Language Models (dLLMs) have emerged as a compelling paradigm for language modeling, offering the unique potential for highly efficient inference via multi-token parallel decoding. However, unlocking this potential remains challenging. High-quality generation typically defaults to one-token-per-step static decoding, while existing parallel algorithms rely on heuristic confidence thresholds, introducing a fragile trade-off between decoding efficiency and generation quality. To overcome this, we introduce Free Draft-and-Verification (FreeDave), a novel training-free and model-free fast decoding algorithm. FreeDave capitalizes on the inherent properties of dLLMs by treating future token predictions as ``free drafts'' and rigorously self-verifying them in subsequent forward passes. Theoretically, FreeDave searches for a shorter decoding path that is guaranteed to reproduce the exact sequence generated by static decoding, achieving algorithmically lossless acceleration. With a sufficiently large draft budget and a verification frontier that does not retreat, it can use the fewest possible model forward passes among feasible paths. Extensive evaluations on math reasoning and code generation benchmarks demonstrate that FreeDave can boost the number of generated tokens per model forward (TPF) by up to 5.48×\times and tokens per second (TPS) by up to 3.94×\times, while robustly maintaining generation quality and avoiding the task-specific performance degradation faced by heuristic methods. Code is available at https://github.com/cychomatica/FreeDave.
Sep 29, 2025

RFG: Self-Improving Diffusion Large Language Models with Reward-Free Guidance

Diffusion Large Language Models (dLLMs) have shown strong reasoning capabilities, yet further improving them typically requires costly post-training with additional data and supervision. We ask whether a post-trained dLLM can improve itself at inference time without additional training, data, or reward models. This requires a guidance signal from the checkpoints alone that is well-defined on the partially masked intermediate states of dLLMs, which existing methods fail to provide. Here we propose Reward-Free Guidance (RFG), a training-free framework for inference-time self-improvement of dLLMs that derives such a signal from the checkpoints themselves. We theoretically demonstrate that a reward signal for a partially masked state can be parameterized by the log-likelihood ratio between a post-trained policy dLLM and its reference model. In practice, this signal can be obtained using off-the-shelf checkpoints alone. Extensive experiments show that RFG consistently improves state-of-the-art post-trained dLLMs by up to 16.1%. Furthermore, despite being entirely training-free, RFG delivers gains that rival or even surpass resource-intensive reinforcement learning techniques.
Jun 24, 2025

Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture

Efficiently scaling Large Language Models (LLMs) necessitates exploring alternatives to dominant autoregressive (AR) methods, with Masked Diffusion Models (MDMs) emerging as candidates. However, comparing AR (typically decoder-only) and MDM (often encoder-only) paradigms is confounded by differing architectures, obscuring true algorithmic and efficiency trade-offs. This research decouples these factors by evaluating MDMs within a decoder-only framework to: (1) Equitably compare MDM (as Any-Order AR) and standard AR paradigms through discrepancies on orders. (2) Investigate MDM architectural impacts on computational efficiency. We show decoder-only MDMs, despite a larger modeling space, can achieve significant inference speedups (∼25×\sim25\times) and comparable perplexity with techniques like temperature annealing, offering a path to reduced inference compute. This work provides insights for developing more computationally efficient foundation models by disentangling core modeling choices from architectural influences. Code is available at https://github.com/scxue/AO-GPT-MDM.