The paper introduces Spatially Speculative Decoding (SSD), an inference-time framework that leverages the two-dimensional correlation of image tokens to accelerate autoregressive image generation. SSD predicts both the adjacent horizontal token and the token directly below it, enabling parallel advancement in multiple spatial directions and reducing backbone forward evaluations. The method achieves up to an 11.03× speedup in wall‑clock time while preserving generation quality on DPG‑Bench and GenEval.
Speculative decoding has significantly accelerated Large Language Model (LLM) inference by alleviating memory-bound bottlenecks. However, traditional speculative decoding typically relies on auxiliary draft modules, incurring significant training and communication overhead.
GroundAnything is a 4‑B parameter grounding foundation model that combines autoregressive and diffusion approaches to achieve fast parallel decoding while maintaining precise visual grounding. By treating grounding as visual evidence extraction and using blockwise denoising, it allows spatial hypotheses to be generated in parallel and refined iteratively. The model outperforms existing state‑of‑the‑art methods on 30 grounding benchmarks, achieving 72.42% accuracy with its autoregressive variant and 61.75% with entropy‑guided decoding, while also offering significant speedups through optional self‑speculative decoding.
The paper introduces a trajectory-level speculative decoding framework for diffusion-based language models (dLLMs), addressing the limitation of existing strategies that revert to single-token generation when confidence is low. By constructing draft denoising trajectories through confidence-stratified tree exploration and verifying them with blockwise parallel evaluation and bidirectional attention masking, the method also incorporates inter-block speculation to exploit the models’ bidirectional structure. Experiments show a 30–40% reduction in denoising iterations, a token-per-step increase from 2.6 to 4.3, and a 7–14× speedup over vanilla dLLMs while maintaining accuracy within 1% on reasoning and code benchmarks.
arXiv:2607. 10661v1 Announce Type: cross Abstract: Speculative decoding has significantly accelerated Large Language Model (LLM) inference by alleviating memory-bound bottlenecks.
arXiv:2609.13245v1 Announce Type: new Abstract: Speculative Jacobi Decoding (SJD) is an important approach for accelerating autoregressive image generation. Although SJD has shown superior performanc...
The paper investigates how attention sparsity behaves in autoregressive image generation, finding a distinct diagonal sparsity pattern due to spatial locality of visual tokens. It introduces a diagonal‑aware sparse attention mechanism that skips KV entries along the diagonal within a recent window, achieving up to 3.1× higher throughput and 1.19× lower latency with less than 2% quality loss compared to dense inference.
arXiv:2609.36590v1 Announce Type: cross Abstract: Self-speculative decoding accelerates large language model (LLM) inference by drafting tokens from the target model itself, but faces a sharp tradeof...
Autoregressive image generation has emerged as a paradigm for multimodal AI systems due to its compatibility with transformer-based LLM serving infrastructures. However, generating thousands of visual...
arXiv:2411. 05894v3 Announce Type: replace-cross Abstract: Speculative Decoding has emerged as a popular technique for accelerating inference in Large Language Models.
arXiv:2606. 00535v1 Announce Type: new Abstract: Speculative decoding (SD) has proven to be an effective technique for accelerating autoregressive generation in large language models (LLMs) however, its application to vision-language models (VLMs) remains relatively unexplored.
arXiv:2606. 02544v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) have recently emerged as a promising alternative to autoregressive (AR) LLMs, offering faster inference through parallel or blockwise decoding.
arXiv:2601. 17917v3 Announce Type: replace Abstract: 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.