{"slug": "how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent", "title": "How would you evaluate a personal AI for invented memories and revoked consent?", "summary": "A practitioner proposal argues that personal AI memory systems should be scored on retrieval and final-answer stages separately, since an invented memory can originate in either, and should be evaluated on abstention alongside accuracy so a system that refuses everything does not appear perfect. The proposal cites LongMemEval (arXiv 2410.10813), which tests information extraction, multi-session reasoning, temporal reasoning, knowledge updates and abstention and splits memory into indexing, retrieval and reading, and TOFU (arXiv 2401.06121), which uses 200 diverse synthetic author profiles with a forget set, as the nearest setup for testing revoked consent by treating a revoked fact as forget-set data and checking it never returns in answers or the store.", "body_md": "Two baselines that cover parts of this:\n\nLongMemEval ( [\\[2410.10813\\] LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://arxiv.org/abs/2410.10813) ) tests “information extraction, multi-session reasoning, temporal reasoning, knowledge updates, and abstention”, and splits memory into “indexing, retrieval, and reading”. That split answers whether to score retrieval and the final answer separately: I think you should, since an invented memory can come from either stage.\n\nFor revoked consent, TOFU ( [\\[2401.06121\\] TOFU: A Task of Fictitious Unlearning for LLMs](https://arxiv.org/abs/2401.06121) ) is the nearest setup: “200 diverse synthetic author profiles” with a “forget set that serves as the target for unlearning”. You can treat a revoked fact as forget-set data and check that it never comes back in answers; not only that, but it is gone from the store.\n\nOne suggestion from my side: score abstention on questions the system should not know, together with accuracy on ones it should. Otherwise, a system that refuses everything looks perfect on invented memories.", "url": "https://wpnews.pro/news/how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent", "canonical_source": "https://discuss.huggingface.co/t/how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent/190672#post_5", "published_at": "2026-10-17 07:29:12+00:00", "updated_at": "2026-10-17 07:32:47.127780+00:00", "lang": "en", "topics": ["ai-safety", "ai-research", "large-language-models", "ai-agents"], "entities": ["LongMemEval", "TOFU", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent", "markdown": "https://wpnews.pro/news/how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent.md", "text": "https://wpnews.pro/news/how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent.txt", "jsonld": "https://wpnews.pro/news/how-would-you-evaluate-a-personal-ai-for-invented-memories-and-revoked-consent.jsonld"}}