Preprint · 2026 · Open access
Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory
arXiv (Cornell University) · Published 30 July 2026 · DOI 10.48550/arxiv.2607.27539
In brief
“Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory” is a 2026 preprint by Vishwajith Ramesh, published in arXiv (Cornell University). OpenAlex counts 3 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).
- Citations
- 3
- Most cited in
- 2026 (3)
- References
- 30
- Authors
- 1
Abstract
An assistant can stop repeating a fact without removing it from memory. To study this difference, we install a support-vector gate in frozen Gemma 3 and record which stored keys and values belong to each exchange. A deletion request excludes the exchange's rows from the long-range readout and recalculates the gate on what remains. We check this operation against an independent refit, then compare it with running the model again on the conversation without the exchange. This second comparison matters because the exchange may already have influenced surviving memory rows. At 4B, the gated model passed checks for recall and feasible deletion on the same six of eight records admitted by the base model, at a perplexity cost under 2%. Admission fell at the smaller and larger checkpoints with the same configuration. The edited memory agreed closely with the local refit on the registered probes, and the model disclosed fewer deleted answers than when simply instructed to forget. However, an attack evaluated separately for each record could still distinguish edited memory from memory that never stored the record. Excluding an exchange's own rows therefore provides a way to edit and audit conversation memory, while leaving a measurable difference from rebuilding it without that exchange. Additional paired studies found no update-speed advantage for the current FP32 proxy and retained-answer matching below half in every tested condition.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Ramesh, V. (2026). Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2607.27539
MLA 9
Ramesh, Vishwajith. “Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2607.27539.
Chicago
Ramesh, Vishwajith. 2026. “Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2607.27539.
Harvard
Ramesh, V. (2026) ‘Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory’, arXiv (Cornell University). doi:10.48550/arxiv.2607.27539.
IEEE
V. Ramesh, “Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2607.27539.
BibTeX
@misc{ramesh2026assistant,
title = {{Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory}},
author = {Ramesh, Vishwajith},
howpublished = {arXiv (Cornell University)},
year = {2026},
doi = {10.48550/arxiv.2607.27539},
url = {https://doi.org/10.48550/arxiv.2607.27539}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- memory editing
- conversational AI assistant
- Gemma 3
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Ramesh, V. (2026). Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2607.27539
How many times has this paper been cited?
3 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Vishwajith Ramesh.
What is the DOI of this paper?
10.48550/arxiv.2607.27539, permanently at https://doi.org/10.48550/arxiv.2607.27539. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.