Metraipolis › archive › dc_ca396e82aadb

ProTTT: Learning to Learn Semantic User Memory with Test-Time Training — arXiv:2609.32564

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32564
License (as archived)
CC0-1.0
Archived
2026-10-07T20:55:43.895Z
SHA-256 of the archived text
231ace1ccbb486ec1975a9821b5ed670a42196c80d74311ec44bbb8b3f5f27ca
Stable URL of this page
https://metraipolis.com/documento/dc_ca396e82aadb

How to cite

ProTTT: Learning to Learn Semantic User Memory with Test-Time Training — arXiv:2609.32564 — arXiv (en), https://arxiv.org/abs/2609.32564 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ca396e82aadb (Metraipolis archive dc_ca396e82aadb, sha256 231ace1ccbb486ec1975a9821b5ed670a42196c80d74311ec44bbb8b3f5f27ca)
BibTeX
@misc{dcca396e82aadb,
  title = {ProTTT: Learning to Learn Semantic User Memory with Test-Time Training — arXiv:2609.32564},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32564},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ca396e82aadb; sha256 231ace1ccbb486ec1975a9821b5ed670a42196c80d74311ec44bbb8b3f5f27ca},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32564. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Personalization requires language models to capture user-specific knowledge from a growing user history. Existing context-based approaches incur increasing inference costs as user history accumulates and rely on separate retrieval or summarization stages, while parametric-based approaches often require reconstructing user representations when new user data is added. We introduce ProTTT, a profile-supervised meta-learning framework for learning semantic user memory. The memory construction starts from a shared initialization and is updated for each user through test-time training on user history, allowing it to evolve continuously as the history grows. However, since test-time training alone does not explicitly encourage the memory to capture semantic user knowledge necessary for personalization, we learn this shared initialization using textual user profiles as supervision, so that test-time training on user history captures semantic knowledge more effectively. ProTTT consistently outperforms both full history ICL and all parametric baselines across diverse benchmarks, while substantially reducing inference cost by compressing user history into a lightweight parameterized memory. Our analysis also shows that profile supervision is a reliable objective for learning semantic user knowledge and that the resulting memory can track and retain evolving user preferences, while remaining robust across different history sizes. Overall, we demonstrate the effectiveness of test-time training for personalization and establish ProTTT as a baseline for continuously evolving user memory.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.32564 — 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 2026-10-07T20:29:00.875Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).