autoregressive models
Models that predict future values based on previously observed values, often used in time series analysis and natural language processing. They generate data sequentially, conditioning each predicted value on past outputs.
- A Unified Framework for Fair Graph Generation: Theoretical Guarantees and Empirical Advances
- Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking
- Aligning Transformers with Continuous Feedback via Energy Rank Alignment
- Anchored Diffusion Language Model
- Architectural and Inferential Inductive Biases for Exchangeable Sequence Modeling
- Bidirectional Representations Augmented Autoregressive Biological Sequence Generation: Application in De Novo Peptide Sequencing
- BitMark: Watermarking Bitwise Autoregressive Image Generative Models
- Breaking AR’s Sampling Bottleneck: Provable Acceleration via Diffusion Language Models
- Constrained Discrete Diffusion
- Context-Aware Regularization with Markovian Integration for Attention-Based Nucleotide Analysis
- Continuous Diffusion Model for Language Modeling
- Corrector Sampling in Language Models
- DINGO: Constrained Inference for Diffusion LLMs
- Diffusion Beats Autoregressive in Data-Constrained Settings
- Edit Flows: Variable Length Discrete Flow Matching with Sequence-Level Edit Operations
- Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning
- Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows
- Forging Time Series with Language: A Large Language Model Approach to Synthetic Data Generation
- Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation
- How Patterns Dictate Learnability in Sequential Data
- LaViDa: A Large Diffusion Model for Vision-Language Understanding
- Large Language Diffusion Models
- Large Language Diffusion Models
- MRO: Enhancing Reasoning in Diffusion Language Models via Multi-Reward Optimization
- NFIG: Multi-Scale Autoregressive Image Generation via Frequency Ordering
- Normalizing Flows are Capable Models for Continuous Control
- OmniGen-AR: AutoRegressive Any-to-Image Generation
- On the Entropy Calibration of Language Models
- Remasking Discrete Diffusion Models with Inference-Time Scaling
- Rendering-Aware Reinforcement Learning for Vector Graphics Generation
- Revisiting Multi-Agent World Modeling from a Diffusion-Inspired Perspective
- Robust Distortion-Free Watermark for Autoregressive Audio Generation Models
- SPMDM: Enhancing Masked Diffusion Models through Simplifing Sampling Path
- STree: Speculative Tree Decoding for Hybrid State Space Models
- Score-informed Neural Operator for Enhancing Ordering-based Causal Discovery
- Selftok-Zero: Reinforcement Learning for Visual Generation via Discrete and Autoregressive Visual Tokens
- SpecMER: Fast Protein Generation with K-mer Guided Speculative Decoding
- Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation
- Towards Better & Faster Autoregressive Image Generation: From the Perspective of Entropy
- Transition Matching: Scalable and Flexible Generative Modeling
- Understand Before You Generate: Self-Guided Training for Autoregressive Image Generation
- Video World Models with Long-term Spatial Memory