Home Knowledge Base dLLMs

dLLMs

No mentions found

This entity hasn't been tracked yet, or Iris is still building its knowledge base.

Related Articles from SNS

dMoE: dLLMs with Learnable Block Experts

arXiv:2605.30876v1 Announce Type: new Abstract: Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive models, offering competitive performance while naturally supporting parallel decoding. However, as dLLMs are increasingly integrated with Mixture-of-Experts (MoE) architectures to scale model capacity, a fundamental mismatch arises between block parallel decoding and token-level expert selection. Specifically, each dLLM forward pass...

arXiv CS 9d ago

dgMARK: Decoding-Guided Watermarking for Diffusion Language Models

arXiv:2601.22985v2 Announce Type: replace Abstract: We propose dgMARK, a decoding-guided watermarking method for discrete diffusion language models (dLLMs). Unlike autoregressive models, dLLMs can generate tokens in arbitrary order. While an ideal conditional predictor would be invariant to this order, practical dLLMs exhibit strong sensitivity to the unmasking order, creating a new channel for watermarking.

arXiv CS 9d ago

Learning Unmasking Policies for Diffusion Language Models

arXiv:2512.09106v4 Announce Type: replace Abstract: Diffusion (Large) Language Models (dLLMs) now match the downstream performance of their autoregressive counterparts on many tasks, while holding the promise of being more efficient during inference. One critical design aspect of dLLMs is the sampling procedure that selects which tokens to unmask at each diffusion step. Indeed, recent work has found that heuristic strategies such as confidence thresholding improve both sample quality and...

arXiv CS 7d ago

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

Announce Type: new Abstract: 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.

arXiv CS 6d ago

SimSD: Simple Speculative Decoding in Diffusion Language Models

Announce Type: new 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. However, their masked language modeling formulation remains incompatible with standard token-level speculative decoding, one of the most effective acceleration techniques for AR models. In AR decoding, the causal mask preserves temporally valid token-level contexts, enabling a...

arXiv CS 8d ago

The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models

Announce Type: replace Abstract: Diffusion Large Language Models (dLLMs) break the rigid left-to-right constraint of traditional LLMs, enabling token generation in arbitrary orders. Intuitively, this flexibility implies a solution space that strictly supersets the fixed autoregressive trajectory, theoretically unlocking superior reasoning potential. However, in this paper, we find that for general reasoning tasks (e.g., mathematics and coding), arbitrary order generation may in fact limit...

arXiv CS 1d ago

dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching

arXiv:2506.06295v3 Announce Type: replace Abstract: Autoregressive Models (ARMs) have long dominated the landscape of Large Language Models. Recently, a new paradigm has emerged in the form of diffusion-based Large Language Models (dLLMs), which generate text by iteratively denoising masked segments. This approach has shown significant advantages and potential.

arXiv CS 1d ago

dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching

arXiv:2506.06295v2 Announce Type: replace Abstract: Autoregressive Models (ARMs) have long dominated the landscape of Large Language Models. Recently, a new paradigm has emerged in the form of diffusion-based Large Language Models (dLLMs), which generate text by iteratively denoising masked segments. This approach has shown significant advantages and potential.

arXiv CS 7d ago

STaR-Quant: State-Time Consistent Post-Training Quantization for Diffusion Large Language Models

arXiv:2606.04945v1 Announce Type: new Abstract: Diffusion large language models (DLLMs) have recently emerged as a promising alternative to autoregressive LLMs by generating text through iterative masked denoising with bidirectional context. However, their large model sizes and iterative denoising process introduce substantial memory and computational overhead, motivating post-training quantization for efficient deployment. In this paper, we identify two key challenges for low-bit DLLM...

arXiv CS 6d ago

STaR-Quant: State-Time Consistent Post-Training Quantization for Diffusion Large Language Models

arXiv:2606.04945v2 Announce Type: replace Abstract: Diffusion large language models (DLLMs) have recently emerged as a promising alternative to autoregressive LLMs by generating text through iterative masked denoising with bidirectional context. However, their large model sizes and iterative denoising process introduce substantial memory and computational overhead, motivating post-training quantization for efficient deployment. In this paper, we identify two key challenges for low-bit DLLM...

arXiv CS 1d ago