← Wednesday, October 21, 2026Models & trainingPaper

Progress-aligned recurrence reaches 95.5% accuracy at 16 times training length

huggingface.co by Pan Wang et al. Editor score 8/10

PAR aligns each recurrent iteration with the state after the same number of operations and normalizes recurrent states; a Qwen2.5-0.5B model reaches 95.5% accuracy at 512 moves after training on at most 32. The ablations identify which training choices enable length extrapolation.

What

Progress-Aligned Recurrence trains each iteration to predict the state after the corresponding number of operations and anchors recurrent-state norms to the first iteration.

Method

On swap-S_5, a model trained through 32 moves reaches 95.5% full-state accuracy at 512 moves; removing norm anchoring cuts extrapolation below 2.2×.

Why it matters

Per-iteration supervision and norm control offer a way to extend recurrent models to sequential problems deeper than those seen in training.

Read on huggingface.co →

Headline, summary, brief and diagram are written by a model from the source's own text; the link, byline and numbers come from the source. Read the original before relying on a claim.