Awesome Diffusion × Autoregression (DiffxAR)

A curated list of hybrid Diffusion + Autoregressive (DiffxAR) models for language, reasoning, and robots.

Note: Currently, this repository is under construction, some tags may be wrong or unclear, and some important papers may not be covered. We welcome everyone's comments and contributions!

---

Legend

Diffusion space

• 🧱 Discrete diffusion (token / categorical space, e.g., D3PM, dLLM).
• 🌊 Continuous / latent diffusion (continuous states or learned latents).

How diffusion × autoregression interact

• 🧩 Planner–executor hybrids (Decompose “thinks / plans” & “speaks / executes”).
• 🪜 Training / objective bridges (convert AR LMs into DLMs, shared objectives, distillation).
• ⚙️ Decoding & efficiency hybrids (blockwise decoding, diffusion-forcing, semi-AR decoding).

Domains

• 🧠 Reasoning & planning (math, logic, tool-use, multi-step CoT).
• 🤖 Embodied / VLA / robotics (vision–language–action, robot policies).
• 🧪 AI4Science & scientific modeling (molecules, materials, proteins, scientific reasoning).
• ⚖️ Surveys & overviews.

---

Paper List

2025

• 2025 🧱🧩🧠 Planner and Executor: Collaboration Between Discrete Diffusion and Autoregressive Models in Reasoning. arXiv:2510.15244
• 2025 🌊🧩🧠 LaDiR: Latent Diffusion Enhances LLMs for Text Reasoning. arXiv:2510.04573
• 2025 🧱🧩 Planned Diffusion: A Guiding Diffusion Language Model via Planning. arXiv:2510.18087
• 2025 🧱🧩🧠 TiDAR: Think in Diffusion, Talk in Autoregression. arXiv:2511.08923
• 2025 🧱🪜⚙️ D2F: Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing. arXiv:2508.09192
• 2025 🧱🧩⚙️ HEX: Test-Time Scaling in Diffusion LLMs via Hidden Semi-Autoregressive Experts. arXiv:2510.05040
• 2025 🧱🪜 SDAR: A Synergistic Diffusion-Autoregression Paradigm for Scalable Sequence Generation. arXiv:2510.06303
• 2025 🧱⚙️ Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models. arXiv:2503.09573
• 2025 🧱⚙️ Sequential Diffusion Language Models. arXiv:2509.24007
• 2025 🧱🪜⚙️ Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive Decoding. arXiv:2508.19529
• 2025 🧱🧠 DiffTOD: Planning with Diffusion Models for Target-Oriented Dialogue Systems. ACL 2025
• 2025 🧱🪜 Energy-Based Diffusion Language Models for Text Generation. arXiv:2410.21357
• 2025 🧱⚖️ On Powerful Ways to Generate: Autoregression, Diffusion, and Beyond. arXiv:2510.06190
• 2025 🧱🪜 Non-Markovian Discrete Diffusion with Causal Language Models. arXiv:2502.09767
• 2025 🌊🧩🤖 HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model. arXiv:2503.10631

2024

• 2024 🧱🧩 Diffusion Guided Language Modeling. arXiv:2408.04220
• 2024 🧱🧠 Beyond Autoregression: Discrete Diffusion for Complex Reasoning and Planning. arXiv:2410.14157
• 2024 🧱⚙️ Beyond Autoregression: Fast LLMs via Self-Distillation Through Time. arXiv:2410.21035
• 2024 🧱🪜 Scaling Diffusion Language Models via Adaptation from Autoregressive Models. arXiv:2410.17891
• 2024 🧱🧠 Diffusion-of-Thought: Chain-of-Thought Reasoning in Diffusion Language Models. arXiv:2402.07754
• 2024 🧱 D3PM: Structured Denoising Diffusion Models in Discrete State-Spaces. arXiv:2107.03006
• 2024 🌊🧩🤖 Diffusion-VLA: Scaling Robot Foundation Models via Unified Diffusion and Autoregression. arXiv:2412.03293

2023

• 2023 🧱🧩 AR-Diffusion: Auto-Regressive Diffusion Model for Text Generation. arXiv:2305.09515
• 2023 🧱 A Reparameterized Discrete Diffusion Model for Text Generation. arXiv:2302.05737
• 2023 🧱 Sequential Data Generation with Groupwise Diffusion Process. arXiv:2310.01400
• 2023 🌊🧩 PLANNER: Enhancing Paragraph Generation with a Latent Language Diffusion Model. arXiv:2306.02531

2022

• 2022 🌊🧩 Latent Diffusion for Language Generation. arXiv:2212.09462
• 2022 🧱 SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Control. arXiv:2210.17432
• 2022 🌊 Diffusion-LM Improves Controllable Text Generation. arXiv:2205.14217
• 2022 🧱 D3PMs: Denoising Diffusion Probabilistic Models for Discrete Data. arXiv:2107.03006

---

Want to Contribute?

We welcome contributions! Please feel free to submit a PR or open an issue if you'd like to add new papers, tools, or correct any mistakes.

✅ Guidelines

• Only add papers that explicitly connect diffusion and autoregression in architecture, training, or decoding.
• Use consistent formatting:
  YEAR ICONS Title. [arXiv:ID](https://arxiv.org/abs/ID)
• Try to tag each paper with:
  • one of 🧱 / 🌊,
  • whether it is 🧩 / 🪜 / ⚙️, and
  • optional domain tags 🧠 / 🤖 / 🧪 / ⚖️.
• Prefer papers with an arXiv entry so that every bullet has a stable link.
• If you add a new domain (e.g. speech, music, code), consider adding a short note in the PR explaining how diffusion and AR interact in that setting.