newsGoogle DeepMindTrust 88 · LabPublished 2mo agoLive · 1mo ago

DiffusionGemma: 4x faster text generation

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

LinkedLinked via unknownDiffusion language models, revisited →
LinkedLinked via unknownminimal-diffusion-lm →
LinkedLinked via unknownDiffusion →
LinkedLinked via unknownDiffuse-XL →
LinkedLinked via unknownSafe Autoregressive Image Generation with Iterative Self-Improving Codebooks →
LinkedLinked via unknownVGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing →
LinkedLinked via unknownAdaptive Block Diffusion: Resolving Training-Inference Mismatch in Diffusion Language Models →
LinkedLinked via unknownMulti-Block Diffusion Language Models →

Covers

paperDiffusion language models, revisited repominimal-diffusion-lm glossary_termDiffusion modelDiffuse-XL

Covers (incoming)

paperSafe Autoregressive Image Generation with Iterative Self-Improving Codebooks paperVGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing paperAdaptive Block Diffusion: Resolving Training-Inference Mismatch in Diffusion Language Models paperMulti-Block Diffusion Language Models paperUnderstanding Evaluation Illusion in Diffusion Large Language Models paperT2LDM++: A Self-Conditioned Representation Guided Diffusion Model for Realistic Text-to-LiDAR Scene Generation paperYour Data Manifold is Secretly a Reward Model: Shell-LCC for Text-to-Video Generation paperIntermediate Text Representation Guided Text-to-Image Generation for Enhancing One-and-Only Alignment paperLittle Brains, Big Feats: Exploring Compact Language Models paperReading Order Inference for Complex Document Layouts paperEquiSteer: Cross-Attention Steering Towards a Fairer Text-Guided Image Generation paperDiffusion-GR2: Diffusion Generative Reasoning Re-ranker repoVCG-team/DiffSegmenter repomanaskng/opp-lens paperAccelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques paperKaleido: Algorithm-Hardware Co-Design for Video Diffusion Transformers by Exploiting Latent Space Correlations paperMulti-Axis Max@K Reinforcement Learning for Representative Diversity in Text-to-Image Generation paperLatent Trajectory Discrimination for AI-Generated Text Detection paperInduction in Both Directions: A Mechanistic Analysis of In-Context Learning in Masked Diffusion Language Models paperAdaptive Multi-Step Lookahead Decoding for Diffusion Language Models paperAppearance Pointers -- Multimodal Region Control of Diffusion Transformers

Related across the graph

paperLatent Trajectory Discrimination for AI-Generated Text Detection paperLittle Brains, Big Feats: Exploring Compact Language Models paperAccelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques paperDiffusion-GR2: Diffusion Generative Reasoning Re-ranker paperInduction in Both Directions: A Mechanistic Analysis of In-Context Learning in Masked Diffusion Language Models paperMulti-Block Diffusion Language Models paperAppearance Pointers -- Multimodal Region Control of Diffusion Transformers paperUnderstanding Evaluation Illusion in Diffusion Large Language Models repomanaskng/opp-lens repominimal-diffusion-lm paperMulti-Axis Max@K Reinforcement Learning for Representative Diversity in Text-to-Image Generation paperKaleido: Algorithm-Hardware Co-Design for Video Diffusion Transformers by Exploiting Latent Space Correlations paperDiffusion language models, revisited paperYour Data Manifold is Secretly a Reward Model: Shell-LCC for Text-to-Video Generation paperIntermediate Text Representation Guided Text-to-Image Generation for Enhancing One-and-Only Alignment paperReading Order Inference for Complex Document Layouts glossary_termDiffusion paperAdaptive Multi-Step Lookahead Decoding for Diffusion Language Models paperAdaptive Block Diffusion: Resolving Training-Inference Mismatch in Diffusion Language Models modelDiffuse-XL repoVCG-team/DiffSegmenter paperEquiSteer: Cross-Attention Steering Towards a Fairer Text-Guided Image Generation paperT2LDM++: A Self-Conditioned Representation Guided Diffusion Model for Realistic Text-to-LiDAR Scene Generation paperSafe Autoregressive Image Generation with Iterative Self-Improving Codebooks paperVGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing