RSPO interprets reward advantages as targets for relative log-ratios in dLLMs, calibrating noisy estimates to stabilize RLVR training and achieve strong gains on planning tasks with competitive math reasoning performance.
hub Canonical reference
LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning
Canonical reference. 78% of citing Pith papers cite this work as background.
25
Pith papers citing it
Background
78% of classified citations
abstract
In this work, we introduce LLaDA-V, a purely diffusion-based Multimodal Large Language Model (MLLM) that integrates visual instruction tuning with masked diffusion models, representing a departure from the autoregressive paradigms dominant in current multimodal approaches. Built upon LLaDA, a representative large language diffusion model, LLaDA-V incorporates a vision encoder and MLP connector that projects visual features into the language embedding space, enabling effective multimodal alignment. Our empirical investigation reveals several intriguing results: First, LLaDA-V demonstrates promising multimodal performance despite its language model being weaker on purely textual tasks than counterparts like LLaMA3-8B and Qwen2-7B. When trained on the same instruction data, LLaDA-V is highly competitive to LLaMA3-V across multimodal tasks with better data scalability. It also narrows the performance gap to Qwen2-VL, suggesting the effectiveness of its architecture for multimodal tasks. Second, LLaDA-V achieves state-of-the-art performance in multimodal understanding compared to existing hybrid autoregressive-diffusion and purely diffusion-based MLLMs. Our findings suggest that large language diffusion models show promise in multimodal contexts and warrant further investigation in future research. Project page and codes: https://ml-gsai.github.io/LLaDA-V-demo/.
hub tools
citation-role summary
background 7
baseline 1
method 1
citation-polarity summary
representative citing papers
GPO-V jailbreaks dVLMs by globally optimizing probabilities in the denoising process to bypass refusal patterns, achieving stealthy and transferable attacks.
BARD bridges autoregressive and diffusion VLMs with progressive block merging plus stage-wise intra-diffusion distillation, delivering 3x speedup and new SOTA on open dVLMs using under 4.4M data points.
DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.
dMLLM-TTS delivers up to 6x more efficient test-time scaling for diffusion MLLMs via O(N+T) hierarchical search and self-verified feedback, improving generation quality on GenEval across three models.
Fast-dLLM adds reusable KV cache blocks and selective parallel decoding to diffusion LLMs, closing most of the speed gap with autoregressive models without retraining.
A masked discrete diffusion model adds token editing at inference and grouped cross-entropy training to reach 0.90 GenEval, 86.9 DPG, and 10.76 HPSv3 scores.
dMoE aggregates token expert distributions to block level in dLLMs, cutting unique experts from 69.5 to 14.6, memory by 76-80%, and latency by 1.14-1.66x while retaining 99.11% performance.
AnyMo is a masked-modeling framework for any-modality human motion generation trained on the new OmniHuMo dataset of 5,000+ hours of multimodal motion sequences.
VRCD prioritizes visually complementary positions during parallel decoding in dMLLMs by measuring attention overlap with the new Visual Redundancy Index, yielding accuracy gains over confidence-based baselines on M^3CoT and MMBench.
Fast-dDrive is a block-diffusion VLA that reports SOTA accuracy on WOD-E2E and nuScenes driving benchmarks together with 12x throughput over autoregressive baselines via section scaffolds and test-time averaging.
Proposes HT-GRPO with sketch-then-paint staged updates, prompt-conditioned importance ratios, and hierarchical credit assignment for dMLLMs, reporting gains on GenEval and DPG plus quality metrics.
Diagnoses mask prior drift and positional attention collapse in LDVLMs and introduces two plug-and-play decoding interventions that raise long-form generation quality without retraining.
MindVLA-U1 is the first unified streaming VLA architecture that surpasses human drivers on WOD-E2E planning metrics while matching VA latency and preserving language interfaces.
ELF applies continuous-time flow matching in embedding space for language generation and reports outperforming prior discrete and continuous diffusion language models with fewer steps.
Cola DLM proposes a hierarchical latent diffusion model that learns a text-to-latent mapping, fits a global semantic prior in continuous space with a block-causal DiT, and performs conditional decoding, establishing latent prior modeling as an alternative to token-level autoregressive language model
Denoising Recursion Models train multi-step noise reversal in looped transformers and outperform the prior Tiny Recursion Model on ARC-AGI.
Stability-Weighted Decoding improves diffusion LLM accuracy by modulating token scores with temporal stability from KL divergence between prediction steps.
Dataset-level metrics in diffusion language models mask substantial sample-level non-determinism that varies with model and system factors, which a new Factor Variance Attribution metric can decompose.
LaDA-Band applies discrete masked diffusion with dual-track conditioning and progressive training to generate vocal-to-accompaniment tracks that improve acoustic authenticity, global coherence, and dynamic orchestration over prior baselines.
Position and step penalty plus visual reasoning guidance fix premature answering and weak visual grounding in diffusion MLLMs, delivering up to 7.5% accuracy gains and over 3x speedup.
Efficient-DLM converts AR models to dLMs via block-wise causal attention and position-dependent masking, yielding higher accuracy and 2.7-4.5x throughput than Dream 7B and Qwen3 4B.
DIA is a training-free method that dynamically adjusts anchor positions in diffusion LLMs to improve format compliance and accuracy on reasoning benchmarks like GSM8K and MATH.
ECHO introduces one-step block diffusion via Direct Conditional Distillation and Response-Asymmetric Diffusion to generate chest X-ray reports faster than autoregressive models while improving clinical metrics.
citing papers explorer
-
Stability-Weighted Decoding for Diffusion Language Models
Stability-Weighted Decoding improves diffusion LLM accuracy by modulating token scores with temporal stability from KL divergence between prediction steps.