Context-Tower Conversion Preserves Generation While Freezing Retains Knowledge: Low-Budget AR-to-Diffusion Conversion of MoE LLMs
Mirrored from arXiv — Machine Learning for archival readability. Support the source by reading on the original site.
Computer Science > Machine Learning
arXiv:2610.02657 (cs)
[Submitted on 2 Oct 2026]
Title:Context-Tower Conversion Preserves Generation While Freezing Retains Knowledge: Low-Budget AR-to-Diffusion Conversion of MoE LLMs
View a PDF of the paper titled Context-Tower Conversion Preserves Generation While Freezing Retains Knowledge: Low-Budget AR-to-Diffusion Conversion of MoE LLMs, by Wentao Lu and 2 other authors
View PDF
HTML (experimental)
Abstract:Converting a pretrained autoregressive (AR) model to a diffusion language model (dLLM) enables parallel generation without pretraining a new model. Published conversion methods differ by roughly three orders of magnitude in training data and have not been compared under a common protocol. We compare two conversions of the same 30B Mixture-of-Experts (MoE) parent, holding the corpus, supervised-token budget, trainable parameter set and evaluation harness fixed, each under its own training recipe. The in-place model updates a subset of the parent's weights using denoising and representation-alignment losses; the frozen-tower model instead conditions through cross-attention on a frozen causal copy of the parent. With 1B training tokens, the frozen-tower model scores 71.60 on HumanEval pass@10 against 6.19 for the in-place model, an 11.6x improvement. At the same budget it also keeps 95% of the parent's GSM8K score and 99% of its MMLU-Pro score. A dense-parent experiment reproduces the HumanEval separation. Within the two-tower design at about 500M tokens, freezing the context tower retains substantially more MMLU-Pro performance than training it, while both give similar observed HumanEval scores. Our theoretical analysis establishes that both conversion classes contain an exact sampler for the AR parent under a hard attention mask and left-to-right commitment of one position per round. Under a shared loss, freezing removes the gradient contribution through the context states. Furthermore, evaluation protocol substantially affects a published 500B-token conversion's scores in both directions across tasks, while its AR parent's scores vary by less than three points, so comparing dLLMs needs a common protocol. These results show that, in the tested low-budget regime, the frozen-tower configuration retains substantially more of the parent's generation performance than in-place conversion.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.02657 [cs.LG] |
| (or arXiv:2610.02657v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02657
arXiv-issued DOI via DataCite (pending registration)
|
Full-text links:
Access Paper:
- View PDF
- HTML (experimental)
- TeX Source
View a PDF of the paper titled Context-Tower Conversion Preserves Generation While Freezing Retains Knowledge: Low-Budget AR-to-Diffusion Conversion of MoE LLMs, by Wentao Lu and 2 other authors
References & Citations
Loading...
Bookmark
Bibliographic Tools
Code, Data, Media
Demos
Related Papers
About arXivLabs
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
IArxiv recommender toggle
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
More from arXiv — Machine Learning
-
Algebraic Integer Quantization for Machine Learning Applications
Oct 12
-
CircuitTree: Composing Circuits into A Syllable Transcriber
Oct 12
-
Behind-the-Meter Photovoltaic Visibility from Net Smart Meters: Deployment Regimes and Operational Insights
Oct 12
-
GlucoPRISM: Protocol-Supervised Pretraining for Cross-Cohort Glucose Transfer
Oct 12
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.