Claimempirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →importance · notableImportance 0.55, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

TopicsMachine Unlearning Large language models Latent Knowledge in Language Models LLM Safety Training

Fine-tuning an RMU-unlearned language model on about ten unrelated examples recovers most hazardous capabilities.

Not yet assessed. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.

Decomposition

This claim has not been assessed yet. Attention goes where its expected value is highest and someone funds it; once this claim's assessment is funded and runs, it may well decompose into subclaims.

or create a grant for this whole area →

Provenance

Where this claim has been said, linked to its canonical form.

1 source · no origin · 1 untraced

1 Untraced

An Adversarial Perspective on Machine Unlearning for AI SafetySep 26, 2024 Untraced · nothing upstream recorded

we show that finetuning on 10 unrelated examples or removing specific directions in the activation space can recover most hazardous capabilities for models edited with RMU, a state-of-the-art unlearning method.

An Adversarial Perspective on Machine Unlearning for AI Safetyextraction 0.85

Abstract, example of adaptive attacks against unlearning.

Cite this claim: a formal citation with its evidence attached

Embed this claim: show it on another site with its current assessment

An embed reports the graph's judgment as it stands when the page is viewed: the status with its confidence, how disputed the claim is, and the date. It links back here.

CardFits its column and sizes itself. Without the script it stays a quotation with its link.

<blockquote class="minerval-claim" data-claim-id="018e8d4b-9585-4914-aa88-3a5efa1207af"><a href="https://minerval.ai/claims/018e8d4b-9585-4914-aa88-3a5efa1207af">Fine-tuning an RMU-unlearned language model on about ten unrelated examples recovers most hazardous capabilities.</a></blockquote>
<script async src="https://minerval.ai/embed.js"></script>

IframeWhere scripts are not allowed. A fixed height; add ?theme=light or ?theme=dark to fix its colours.

<iframe src="https://minerval.ai/embed/claims/018e8d4b-9585-4914-aa88-3a5efa1207af" title="Minerval: Fine-tuning an RMU-unlearned language model on about ten unrelated examples recovers most…" width="480" height="225" style="border:0;max-width:100%" loading="lazy"></iframe>

BadgeAn image for a README or a forum post, in Markdown.

[![Minerval's judgment of this claim](https://minerval.ai/embed/claims/018e8d4b-9585-4914-aa88-3a5efa1207af/badge.svg)](https://minerval.ai/claims/018e8d4b-9585-4914-aa88-3a5efa1207af)

The badge now:

Contribute

Every judgment on this page is open to challenge. A contribution is evaluated on its merits by the reviewer; if it succeeds the page changes, and if it does not, the reasons are stated. Either way the exchange becomes part of the claim’s public record.

Created by extractor · Nov 11, 2026. Every judgment on this page is accompanied by a reasoning trace.