Metraipolis › archive › dc_204cebaeb79e
Test-Time Adaptation of Reasoning Strategies with Bayesian Nonparametric Memory — arXiv:2610.06516
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.06516
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:28:34.507Z
- SHA-256 of the archived text
953da1f48efe72b9993d6ebfbc86090020d582c774531107aad665f01d105e1e- Stable URL of this page
https://metraipolis.com/documento/dc_204cebaeb79e
How to cite
Test-Time Adaptation of Reasoning Strategies with Bayesian Nonparametric Memory — arXiv:2610.06516 — arXiv, https://arxiv.org/abs/2610.06516 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_204cebaeb79e (Metraipolis archive dc_204cebaeb79e, sha256 953da1f48efe72b9993d6ebfbc86090020d582c774531107aad665f01d105e1e)
BibTeX
@misc{dc204cebaeb79e,
title = {Test-Time Adaptation of Reasoning Strategies with Bayesian Nonparametric Memory — arXiv:2610.06516},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.06516},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_204cebaeb79e; sha256 953da1f48efe72b9993d6ebfbc86090020d582c774531107aad665f01d105e1e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.06516. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
While modern large language models (LLMs) have been trained to reason through verbalized chains-of-thought, the generation cost grows substantially due to suboptimal paths to reach the final answer. Furthermore, as new insights are discovered while observing various input queries (e.g. through self-reflection), limited mechanisms exist for carrying forward these findings to be applied to subsequent problems. One can view the list of such strategies or behaviors as a growing cheatsheet, with elements retrieved from this memory module at inference-time. In this work, we consider structured cheatsheets, with learned clusters of behaviors. We introduce a Hierarchical Dirichlet Process Gaussian Mixture Model (HDP-GMM) over behavior embeddings, which shares components across domains while allowing domain-specific mixing weights, and uses the posterior predictive to retrieve relevant behaviors for a query; we call this a $\textit{Bayesian Cheatsheet}$. This mechanism allows for cheap adaptation in an online test-time training (TTT) setting, softly updating the mixture's sufficient statistics following each sample and enabling the creation of new components when the synthesized behaviors are sufficiently novel. We demonstrate that Bayesian Cheatsheet achieves clear performance gains relative to existing memory modules across reasoning benchmarks such as AIME'25, Omni-MATH, and PhysReason, even in the cold-start setting. We show that the Bayesian Cheatsheet is an adaptively reorganizing memory module, as behaviors can be re-assigned to components through a single step of collapsed Gibbs sampling. Our findings highlight the value of Bayesian-inspired memory modules for effective test-time adaptation and the role of structure in metacognitive reasoning.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.06516 — 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 2026-10-06T05:24:20.375Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).