Metraipolis › archive › dc_ae86a692b07c

MemShift: Separating Detection from Repair in LLM-Derived Personal Memory — arXiv:2611.02338

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02338
License (as archived)
CC0-1.0
Archived
2026-11-03T11:07:37.328Z
SHA-256 of the archived text
7ea23cdb464e1f7c2dfdb655937707c91789517808ed530d263e2d7e9d20c1e9
Stable URL of this page
https://metraipolis.com/documento/dc_ae86a692b07c

How to cite

MemShift: Separating Detection from Repair in LLM-Derived Personal Memory — arXiv:2611.02338 — arXiv (en), https://arxiv.org/abs/2611.02338 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ae86a692b07c (Metraipolis archive dc_ae86a692b07c, sha256 7ea23cdb464e1f7c2dfdb655937707c91789517808ed530d263e2d7e9d20c1e9)
BibTeX
@misc{dcae86a692b07c,
  title = {MemShift: Separating Detection from Repair in LLM-Derived Personal Memory — arXiv:2611.02338},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02338},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ae86a692b07c; sha256 7ea23cdb464e1f7c2dfdb655937707c91789517808ed530d263e2d7e9d20c1e9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02338. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

AI assistants store values that a language model derived from a person's records under the person's own definitions, such as their important contacts. Keeping such a value correct takes two steps: detection, noticing that an input changed, and repair, producing the correct new value. Benchmarks that score only the final answer cannot tell which step failed, and they state every change as new text. Seeing this limitation, we present MemShift, a benchmark of five shifts: source revision, deletion, redefinition, writer switch, and export. The first three fork a synthetic world into branches that differ only in the shift. Gold values and dependencies are fixed before any text, and a deterministic scorer replaces an LLM judge. For a memory that reports its invalidations, MemShift shows for each changed item whether the memory noticed the change and whether its repair was right. Moreover, we build a reference memory with known detection and repair to validate the measure. By evaluating four open-weight LLMs, we find that detection helps only as much as repair allows: the LLMs re-derive few values computed by the person's definitions correctly, and programs that the LLM writes over typed records close most of the gap to exact recomputation. Code and data are available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.02338 — 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 2026-11-03T10:59:46.091Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).