{"schemaVersion":1,"target":{"type":"reading","id":"3d7af2f47fe4a0fc8382fefff45bef1dcea6467b5705c4b10a01e9727d75023e","revision":"c43f9020-934a-4618-828e-a563db8a7f9c","revisionKind":"publication-id","title":"arXiv cs.AI: Addition by Subtraction: Toward Mitigating Fragile Safety Alignment in Large Language Models","url":"https://escapemeridian.com/readings/arxiv-cs-ai-addition-by-subtraction-toward-mitigating-fragile-3d7af2f4"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.06553","revision":"bdcab4dd0cb00071c469c5c565b2d05a02aa7c3216b1d11896f68786ce1cfcde","revisionKind":"sha256-source-text","title":"Addition by Subtraction: Toward Mitigating Fragile Safety Alignment in Large Language Models","url":"https://arxiv.org/abs/2611.06553"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."}