Metraipolis › archive › dc_e40f6caeb42f
RecurTrace: Adaptive Latent Reasoning with Loop-Time Memory — arXiv:2609.03379
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.03379
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T02:07:52.999Z
- SHA-256 of the archived text
65b13aff19121cbe15ec03a039810ade30fa5d298360def5752d07dbd6e97260- Stable URL of this page
https://metraipolis.com/documento/dc_e40f6caeb42f
How to cite
RecurTrace: Adaptive Latent Reasoning with Loop-Time Memory — arXiv:2609.03379 — arXiv (en), https://arxiv.org/abs/2609.03379 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e40f6caeb42f (Metraipolis archive dc_e40f6caeb42f, sha256 65b13aff19121cbe15ec03a039810ade30fa5d298360def5752d07dbd6e97260)
BibTeX
@misc{dce40f6caeb42f,
title = {RecurTrace: Adaptive Latent Reasoning with Loop-Time Memory — arXiv:2609.03379},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.03379},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e40f6caeb42f; sha256 65b13aff19121cbe15ec03a039810ade30fa5d298360def5752d07dbd6e97260},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03379. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Repeating a small block of middle layers increases a language model's effective inference depth without adding parameters or generating extra tokens, and recent work shows that this latent recurrence improves reasoning. However, two design choices limit these gains. Each iteration sees only the previous output and cannot directly access earlier computations. Moreover, a fixed loop count wastes depth on easy inputs while leaving hard ones with too little computation. We introduce RecurTrace, which addresses both limitations using the loop's own trajectory. Specifically, Loop Memory Attention lets each looped layer attend to its own states from previous iterations along the loop-time axis, so the model can revisit earlier computations instead of relying on the latest state alone. A halting head then reads the loop state and predicts whether to continue, with supervision from an oracle that identifies when additional depth still reduces loss. In a controlled MathQA comparison on the same looped backbone, RecurTrace achieves 56.9% accuracy with an average of 2.0 loops, exceeding the best fixed loop depth by 2.2 points at matched compute. By comparison, ACT and PonderNet collapse to one loop, and CALM reaches only 54.1% with 5.6 loops, while the stronger LoopUS-Conf and TaH-Mismatch baselines reach 55.3% at 3.2 loops and 55.7% at 2.1 loops. Finally, RecurTrace improves generation accuracy over same-budget fine-tuned baselines at 0.6B, 1.7B, 4B, and 8B, with the gain growing with model size from 0.6 to 3.4 points. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.03379 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T01:47:25.544Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).