Efficient Pretraining Length Scaling
Adds hidden decoding tokens during pretraining, then discards their key/value cache entries after use. Keeps the original tokens' entries to preserve longer-range context.
First public:
Cite this paper
ArXiv preprint
arXiv:2504.14992 · Initial submission: 2025-04-21
A published-version citation has not been verified for this entry. This does not mean the paper is unpublished.
Choose the version you used. The arXiv link may serve a revision; check the original paper for the exact result, assumptions and limitations you cite.
Explore this approach
Thinking tokens — Give a model extra tokens for computation, such as pauses, filler symbols, or learned planning tokens. These tokens need not spell out the solution.
Recurrent reasoning — Repeat parts of the model or keep intermediate information in memory while working through a problem. Some methods also learn how many passes to run.
Read or reuse this entry
Plain-text Markdown · JSON record · Find in the full list
Our summary and categories are a reading aid, not the authors’ abstract. Use the original paper for research claims. Suggest a correction.