mlsanityData to 7 Oct 2026

Stable-DiffCoder: Pushing the Frontier of Code Diffusion Large Language Model

arXiv 2601.15892 ↗First posted 22 Jan 2026cs.CLcited by 12

Not on the map. Why?

Abstract (arXiv)

Diffusion-based language models (DLLMs) offer non-sequential, block-wise generation and richer data reuse compared to autoregressive (AR) models, but existing code DLLMs still lag behind strong AR baselines under comparable budgets. We revisit this setting in a controlled study and introduce Stable-DiffCoder, a block diffusion code model that reuses the Seed-Coder architecture, data, and training pipeline. To enable efficient knowledge learning and stable training, we incorporate a block diffusion continual pretraining (CPT) stage enhanced by a tailored warmup and block-wise clipped noise schedule. Under the same data and architecture, Stable-DiffCoder overall outperforms its AR counterpart on a broad suite of code benchmarks. Moreover, relying only on the CPT and supervised fine-tuning stages, Stable-DiffCoder achieves stronger performance than a wide range of ~8B ARs and DLLMs, demonstrating that diffusion-based training can improve code modeling quality beyond AR training alone. Moreover, diffusion-based any-order modeling improves structured code modeling for editing and reasoning, and through data augmentation, benefits low-resource coding languages.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 55 of its 64 references that are in the map. First posted on arXiv 22 Jan 2026.

Builds on

55 of 64 references.

References · 55

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2024The Llama 3 Herd of Modelscited by 14,192
  4. 2025Qwen3 Technical Reportcited by 8,195
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  6. 2021Program Synthesis with Large Language Modelscited by 3,509
  7. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
  8. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  9. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,332
  10. 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,154
  11. 2024Qwen2.5-Coder Technical Reportcited by 1,151
  12. 2025Large Language Diffusion Modelscited by 872
38 more
  1. 2024Simple and Effective Masked Diffusion Language Modelscited by 800
  2. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distributioncited by 712
  3. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 606
  4. 2025Dream 7B: Diffusion Large Language Modelscited by 515
  5. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
  6. 2025Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Modelscited by 385
  7. 2024Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Datacited by 371
  8. 2024DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligencecited by 325
  9. 2025LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Modelscited by 273
  10. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Modelscited by 268
  11. 2025Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusionscited by 208
  12. 2025DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generationcited by 184
  13. 2025Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inferencecited by 173
  14. 2022Multi-lingual Evaluation of Code Generation Modelscited by 158
  15. 2025LLaDA2.0: Scaling Up Diffusion Language Models to 100Bcited by 154
  16. 2022Concrete Score Matching: Generalized Score Matching for Discrete Datacited by 144
  17. 2025Fast-dLLM v2: Efficient Block-Diffusion LLMcited by 128
  18. 2024OpenCoder: The Open Cookbook for Top-Tier Code Large Language Modelscited by 127
  19. 2025SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generationcited by 118
  20. 2025Dream-Coder 7B: An Open Diffusion Language Model for Codecited by 81
  21. 2024Heavy-Tailed Class Imbalance and Why Adam Outperforms Gradient Descent on Language Modelscited by 78
  22. 2014Making Risk Minimization Tolerant to Label Noisecited by 54
  23. 2025LLaDA-MoE: A Sparse MoE Diffusion Language Modelcited by 54
  24. 2023Can It Edit? Evaluating the Ability of Large Language Models to Follow Code Editing Instructionscited by 46
  25. 2025Seed-Coder: Let the Code Model Curate Data for Itselfcited by 46
  26. 2025Diffusion Language Models are Super Data Learnerscited by 45
  27. 2025ParallelBench: Understanding the Trade-offs of Parallel Decoding in Diffusion LLMscited by 42
  28. 2025WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inferencecited by 38
  29. 2025TiDAR: Think in Diffusion, Talk in Autoregressioncited by 35
  30. 2025dInfer: An Efficient Inference Framework for Diffusion Language Modelscited by 33
  31. 2025Scaling Behavior of Discrete Diffusion Language Modelscited by 32
  32. 2025From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMscited by 22
  33. 2025Training Optimal Large Diffusion Language Modelscited by 21
  34. 2024NaturalCodeBench: Examining Coding Performance Mismatch on HumanEval and Natural User Promptscited by 19
  35. 2025Sequential Diffusion Language Modelscited by 19
  36. 2025LoPA: Scaling dLLM Inference via Lookahead Parallel Decodingcited by 17
  37. 2025Exploring the Potential of Diffusion Large Language Models in Code Generationcited by 16
  38. 2025FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Modelscited by 16

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 12.

12 papers.

  1. 2026DMax: Aggressive Parallel Decoding for dLLMscited by 14
  2. 2026Stop Training for the Worst: Progressive Unmasking Accelerates Masked Diffusion Trainingcited by 8
  3. 2026DLM-Scope: Mechanistic Interpretability of Diffusion Language Models via Sparse Autoencoderscited by 6
  4. 2026UnMaskFork: Test-Time Scaling for Masked Diffusion via Deterministic Action Branchingcited by 2
  5. 2026LFPO: Likelihood-Free Policy Optimization for Masked Diffusion Modelscited by 2
  6. 2026TACG: Trajectory-Aware Commit Gating for Diffusion Language Model Decodingcited by 2
  7. 2026AR-MAP: Are Autoregressive Large Language Models Implicit Teachers for Diffusion Large Language Models?cited by 1
  8. 2026DICE: Diffusion Large Language Models Excel at Generating CUDA Kernelscited by 1
  9. 2026Discrete Tilt Matchingcited by 0
  10. 2026Understanding and Accelerating the Training of Masked Diffusion Language Modelscited by 0
  11. 2026Optimality of FSQ Tokens for Continuous Diffusion for Categorical Data with Application to Text-to-Speechcited by 0
  12. 2026How Should Diffusion Language Models Edit Code?cited by 0