About this paper
LoopFill reduces decoding latency for looped language models by combining multi-token suffix drafts and shallow-loop predictions into a single draft tree with full-depth verification that preserves the autoregressive output distribution.
Looped Transformers improve parameter efficiency by repeatedly applying the same Transformer layers. During autoregressive decoding, however, generating each token requires several sequential passes through these layers, leading to high decoding latency. Pipelined self-speculative methods use shallow-loop predictions to process tokens at different loops in the same batch, but each token must reach the drafting depth before the next can enter the pipeline. We introduce LoopFill, a training-free speculative decoding algorithm for looped LLM that combines multi-token suffix drafts with shallow-loop predictions in a single draft tree. Both drafters contribute to one tree, so a suffix lookup…
Looped Transformersparameter efficiencyTransformer layersautoregressive decodingdecoding latencypipelined self-speculative methodsshallow-loop predictionsspeculative decodinglooped LLMmulti-token suffix drafts
- Upvotes
- 4
- Authors
- 4
- Published
- Nov 14, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Trending history
LoopFill: Pipelined Multi-Token Speculative Decoding for Looped Transformers was first seen on Hugging Face · Daily papers on Nov 17, 2026. Since then it has appeared in 4 checks on 2 different days. Its best position was #26 of 30, reached on Nov 17, 21:28 UTC. It dropped off the list after Nov 18, 01:28 UTC.