Prune-then-Loop: Turning a Pretrained VLM into a Recurrent Model
HildeKuehne · x · 2026-10-31
A new paper introduces a prune-then-loop strategy: instead of training from scratch or adding a vision backbone to a recurrent LLM, the authors take a pretrained VLM, remove a subset of its language-model layers, and reuse the remaining layers multiple times, with a linear adapter connecting the recurrent blocks—converting an existing pretrained VLM into a recurrent model. Paper and project page are available.
Related event: Prune-then-Loop Retrofitting Turns Pretrained VLMs into Recurrent Models(3 posts)→
More from Models
- Rumor: OpenAI delayed new model over alignment issues, more than any other lab — haider1 · 2026-10-31
- Google's Argon complains it must verify facts but can't search on its own — Avg_SD_enjoyer · 2026-10-31
- ARC-AGI Kaggle competitions hit major slowdowns as ARC Prize refuses deadline extension — JFPuget · 2026-10-31
- Speech model taxonomy: native S2S and full-duplex are independent dimensions, not exclusive categories — rdesh26 · 2026-10-31
- Pre-registered training-curve forecast hits: TabPFN covers 14/14 later checkpoints — ZimingLiu11 · 2026-10-31
- KAN author argues training-curve prediction fits tabular foundation models like TabPFN, not LLMs — ZimingLiu11 · 2026-10-31