Metraipolis › archive › dc_d9fe41ef03bc

Handover of In-Context Learning State Across Session Boundaries — arXiv:2608.14528

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.14528
License (as archived): CC0-1.0
Archived: 2026-10-19T02:20:53.636Z
SHA-256 of the archived text: 24d1fd0d6a010d71962022ee9fa3b7f4301686359485305086e2f90a0b4dbef9
Stable URL of this page: https://metraipolis.com/documento/dc_d9fe41ef03bc

How to cite

Handover of In-Context Learning State Across Session Boundaries — arXiv:2608.14528 — arXiv (en), https://arxiv.org/abs/2608.14528 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d9fe41ef03bc (Metraipolis archive dc_d9fe41ef03bc, sha256 24d1fd0d6a010d71962022ee9fa3b7f4301686359485305086e2f90a0b4dbef9)

BibTeX

@misc{dcd9fe41ef03bc,
  title = {Handover of In-Context Learning State Across Session Boundaries — arXiv:2608.14528},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14528},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d9fe41ef03bc; sha256 24d1fd0d6a010d71962022ee9fa3b7f4301686359485305086e2f90a0b4dbef9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14528. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This study investigates the methodological and theoretical properties of session handover in applications that use large language models. A task may continue in a new session when the context reaches the model's input limit, when the application restarts, or when another agent is asked to finish the task. The application must then decide which information from the earlier session to pass on. We formulate handover as the transfer of a task-relative in-context learning (ICL) state and distinguish exact recovery of earlier material from preservation of the target distribution. Under an exogeneity condition, predictive equivalence characterizes the coarsest deterministic sufficient handover and gives a fixed-length bit requirement. The analysis isolates the effects of the memory constraint, the writer, and the continuation procedure, and quantifies the cost of writing before the realized downstream query is known. We propose a three-part record that stores decisions and constraints exactly, uses task-justified statistics for repeated evidence, and retains original observations whose effect is not preserved by those statistics. Gaussian linear regression gives an exact finite-dimensional handover and finite-bit perturbation bounds, while nonparametric regression gives upper and lower bounds that relate memory to squared prediction error. These results provide a theory and method for deciding what a handover must retain and how its memory requirement depends on the continuation task.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Econometrics (econ.EM); Statistics Theory (math.ST); Methodology (stat.ME); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2608.14528 — 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 2026-10-19T02:02:02.831Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).