Metraipolis › archive › dc_1fbcfd2e3202
Exact Record Omission in Delta Attention: A Transport Criterion, Its Cost, and a Replay Certificate — arXiv:2609.06872
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.06872
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T12:57:00.448Z
- SHA-256 of the archived text
635d84a4cc73dc738f9f842ced548da45998bef0338ee3ab13c9140ed9a89538- Stable URL of this page
https://metraipolis.com/documento/dc_1fbcfd2e3202
How to cite
Exact Record Omission in Delta Attention: A Transport Criterion, Its Cost, and a Replay Certificate — arXiv:2609.06872 — arXiv (en), https://arxiv.org/abs/2609.06872 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1fbcfd2e3202 (Metraipolis archive dc_1fbcfd2e3202, sha256 635d84a4cc73dc738f9f842ced548da45998bef0338ee3ab13c9140ed9a89538)
BibTeX
@misc{dc1fbcfd2e3202,
title = {Exact Record Omission in Delta Attention: A Transport Criterion, Its Cost, and a Replay Certificate — arXiv:2609.06872},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.06872},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1fbcfd2e3202; sha256 635d84a4cc73dc738f9f842ced548da45998bef0338ee3ab13c9140ed9a89538},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06872. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
An assistant can stop repeating a deleted statement while its recurrent memory still carries that statement's influence. We examine this distinction by saving the state difference immediately after a record, carrying this saved difference, or receipt, through later updates, and comparing the corrected state with the state built from the same conversation with the record omitted. Unrolling the recurrence gives an exact criterion: transport reaches this never-stored state if and only if the additional differences created by later updates cancel after transport. We evaluate eleven conditions using 40 prospectively selected synthetic Kimi Linear contexts, with 40 separate calibration contexts and paired uncertainty estimates computed across complete records. Attention masking removed all 40 exact greedy target answers, yet a fixed three-query candidate-scoring attack with known candidate values and log-probability access attained AUC 0.740 (95% interval 0.695--0.809); the measured false-positive rate was 7.5%. Prompt-only forgetting still returned 33 of the 40 targets. Checkpoint replay matched the complete declared active state and audit logits in all 80 contexts, while preserving all 120 retained answers in the evaluation cohort. An independent Qwen cohort and matched original-bf16/8-bit Kimi controls confirmed native transport mismatch beyond the numerical error measured in matched controls. Replay also supported record replacement and successive deletions with a new record added between them. These results provide a practical audit of both the requested memory change and the assistant's remaining behavior, with replay work determined by the surviving suffix. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL). Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL); Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.06872 — 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 2026-10-16T12:30:25.294Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).