# BYOD: Build Your Own DLM - Efficient Autoregressive-to-Diffusion Language Model Conversion

**Authors:** Ruurd J. A. Kuiper, Robert A. Bagheri  
**Lab:** University Medical Center Utrecht · **Published:** 29 Oct 2026 · **Upvotes:** 8  
**Links:** [arXiv](https://arxiv.org/abs/2610.41574) · [PDF](https://arxiv.org/pdf/2610.41574) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2610.41574.html)

## Key numbers

- **50.6%**: reaching 50.6% average accuracy compared with 51.9% for LLaDA 8B Instruct in the same harness
- **51.9%**: reaching 50.6% average accuracy compared with 51.9% for LLaDA 8B Instruct in the same harness
- **102.4M**: each 25,000-iteration conversion needs at most 102.4M tokens

## Abstract

Unlike autoregressive (AR) models, diffusion language models (DLMs) permit parallel token updates, flexible generation, and bidirectional attention, but leading systems rely on costly pretraining or full-model fine-tuning. We present Build Your Own DLM (BYOD), a generalizable LoRA-only recipe for parameter-efficient AR-to-DLM conversion. Without architecture-specific tuning, BYOD induces diffusion generation in Gemma 2 9B, Llama 3.1 8B, Qwen2.5 7B, and Ministral 8B; to our knowledge, this is the first reported AR-to-DLM conversion of a Mistral model. The adapters comprise 4.1-5.8% of final parameters while every base-model parameter remains frozen; each 25,000-iteration conversion needs at most 102.4M tokens, processed in 3.1-4.0 hours on one Google Colab GPU. All four models generate from fully masked sequences, use bidirectional context in a controlled attention test, and retain substantial parent-model knowledge. BYOD-Gemma is the strongest BYOD conversion on our nine-task evaluation, reaching 50.6% average accuracy compared with 51.9% for LLaDA 8B Instruct in the same harness. Open-generation quality at low network function evaluation (NFE) budgets remains the principal limitation: BYOD perplexity is 8.3-11.1 versus 4.1 for LLaDA and 2.2-3.0 for the AR parents when generating 128 tokens in 64 steps. To enable further experimentation and inspection, we provide a standalone training and evaluation implementation and interactive demos for all four conversions. BYOD thus provides a practical route for extending diffusion generation to other existing model families while identifying the gap that efficient conversion must next close.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.