arXiv:2609.03528v1 Announce Type: cross Abstract: Diffusion reinforcement learning (RL) has recently achieved significant success in post-training image and video generative models. However, most diffusion RL methods, including DanceGRPO and FlowGRPO, recompute selected timesteps…
Diffusion reinforcement learning (RL) has recently achieved significant success in post-training image and video generative models. However, most diffusion RL methods, including DanceGRPO and FlowGRPO, recompute selected timesteps with gradient tracking after rollout. Under on-po…
arXiv:2609.02108v1 Announce Type: cross Abstract: Diffusion language models (DLMs) have emerged as a promising alternative to the auto-regressive paradigm. With bidirectional attention and any-order generation, DLMs naturally fit infilling tasks, which require generating a middle…
arXiv:2609.00873v1 Announce Type: new Abstract: Diffusion language models (DLMs) have recently emerged as an alternative modeling paradigm to autoregressive LMs, offering advantages such as parallel generation and bidirectional context modeling. Despite growing interest in their …
arXiv cs.AI
TIER_1English(EN)·Guoli Wang, Haonan Shi, Tu Ouyang, An Wang·
arXiv:2609.00495v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) generate text through iterative denoising rather than left-to-right decoding. This generation paradigm introduces two axes that can influence safety alignment: when tokens are generated duri…
arXiv cs.AI
TIER_1English(EN)·Yang Li, Han Meng, Chenan Wang, Zhenyu Bi, Xuan Wang, Haipeng Chen·
arXiv:2601.03199v2 Announce Type: replace-cross Abstract: Diffusion language models (DLMs) have shown strong potential for general natural language tasks with in-context examples. Existing In-Context Learning (ICL) approaches largely inherit the practice of autoregressive languag…
arXiv cs.AI
TIER_1English(EN)·Wail Bouhedja, Amr Mohamed, Guokan Shang·
arXiv:2608.30922v1 Announce Type: new Abstract: Masked diffusion language models predict tokens from a partially observed response canvas, enabling bidirectional conditioning and parallel token refinement. Yet standard masked-diffusion decoders use a rigid inference interface: th…
arXiv cs.AI
TIER_1English(EN)·Wenxuan Guo, Yuyang Hong, Lubin Fan, Zhaojin Fu, Lin Chen, Kun Ding, Shiming Xiang·
Existing approaches for synthesizing Partial Differential Equation (PDE) solvers predominantly rely on autoregressive models, yet their global left-to-right decoding incurs substantial redundancy when addressing inherently localized bugs. In this work, we challenge this inefficie…
arXiv cs.CL
TIER_1English(EN)·Kaihua Liang, An Zhong, Xin Tan, Zafar Ayyub Qazi, Hong Xu, Jian Weng, Marco Canini·
arXiv:2608.26140v1 Announce Type: new Abstract: Diffusion Large Language Models (DLLMs) enable non-autoregressive decoding and bidirectional context modeling, but efficient inference remains challenging. Unlike autoregressive systems, whose key-value (KV) cache can be reused for …
arXiv:2608.26574v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) offer a promising alternative to autoregressive generation by decoding multiple tokens in parallel through iterative denoising. However, increasing decoding parallelism often degrades generati…
arXiv cs.CL
TIER_1English(EN)·Haotian Sun, Rushi Qiang, Yuqian Zheng, Bo Dai·
arXiv:2606.08357v2 Announce Type: replace Abstract: Diffusion language models generate text through iterative denoising, offering a powerful alternative to autoregressive generation. However, discrete language spaces lack a natural neighborhood structure for defining effective pe…
arXiv cs.CL
TIER_1English(EN)·Ivan Kobyzev, Abbas Ghaddar, Yufei Cui·
arXiv:2608.26374v1 Announce Type: new Abstract: Diffusion language models (DLMs) generate text by iteratively denoising masked sequences, but standard decoding either fixes the sequence length or relies on ad hoc stopping rules, often leading to unnecessary denoising steps. We re…
arXiv:2608.25311v1 Announce Type: new Abstract: Diffusion Language Models (DLMs) have recently become increasingly competitive with autoregressive (AR) models, and even outperform them on certain tasks. Unlike AR models, DLMs produce output through iterative denoising without a l…
Diffusion Language Models (DLMs) have recently become increasingly competitive with autoregressive (AR) models, and even outperform them on certain tasks. Unlike AR models, DLMs produce output through iterative denoising without a left-to-right order. To further improve the perfo…
arXiv:2601.23182v2 Announce Type: replace Abstract: Despite the non-autoregressive potential of diffusion language models (dLLMs), existing decoding strategies demonstrate positional bias, failing to fully unlock the potential of arbitrary generation. In this work, we delve into …
arXiv:2510.01028v2 Announce Type: replace Abstract: Large language models have made revolutionary progress in generating human-like text, yet their outputs often tend to be generic, exhibiting insufficient structural diversity, which limits personalized expression. Recent advance…
arXiv cs.AI
TIER_1English(EN)·Farhana Amin, Sabiha Afroz, Dimitrios S. Nikolopoulos·
arXiv:2608.22646v1 Announce Type: new Abstract: Diffusion language models can generate many tokens in parallel, but they still require repeated denoising steps during inference. This makes generation costly, especially when the model continues to recompute tokens that are already…
arXiv:2608.23167v1 Announce Type: new Abstract: Diffusion Language Models (DLMs) exhibit strong parallel decoding capabilities by denoising multiple tokens in a single generation step. However, this parallelism comes with substantial computational overhead, as each step requires …
arXiv cs.CL
TIER_1English(EN)·Hyeongsoo Lim, Jinyoung Kim, Eunseo Seo, Minho Jang, Jiwon Yoon·
arXiv:2608.22898v1 Announce Type: new Abstract: Diffusion language models (DLMs) alleviate the inherent latency bottleneck of autoregressive (AR) large language models (LLMs), but their degraded generation quality limits practical applicability. Although knowledge distillation (K…
Diffusion Language Models (DLMs) exhibit strong parallel decoding capabilities by denoising multiple tokens in a single generation step. However, this parallelism comes with substantial computational overhead, as each step requires interactions with all suffix tokens. Existing me…
arXiv:2603.02760v2 Announce Type: replace-cross Abstract: Diffusion large language models (dLLMs) have recently attracted significant attention for their ability to enhance diversity, controllability, and parallelism. However, their non-sequential, bidirectionally masked generati…
arXiv:2506.00290v2 Announce Type: replace-cross Abstract: This paper introduces DLM-One, a score-distillation-based framework for one-step sequence generation with continuous diffusion language models (DLMs). DLM-One eliminates iterative refinement by aligning the scores of a stu…
arXiv:2609.01043v1 Announce Type: cross Abstract: Uniform-state discrete diffusion models update all tokens in parallel while keeping every position revisable. Even when the commonly used top-$p$ rule leaves only one candidate at a position, that choice affects only the current r…
arXiv:2505.16990v3 Announce Type: replace Abstract: In this work, we propose Dimple, the first Discrete Diffusion Multimodal Large Language Model (DMLLM). We observe that training with a purely discrete diffusion approach leads to significant training instability, suboptimal perf…
Hacker News — AI stories ≥50 points
TIER_1English(EN)·peter_d_sherman·
<p>If you have watched an AI write, you know the ritual. Tokens appear left to right, one after another, like someone typing very fast. It feels like proof of intelligence. It is actually a constraint. Every mainstream language model, from GPT to Claude to the small model running…
🧠 Researchers introduce Continuous Diffusion Language Models, which apply diffusion processes to generate text by iteratively refining token representations in continuous space. The approach differs from traditional autoregressive language models by using a continuous refinement …
