Study Rethinks Supervision Targets for Multimodal LLM Unlearning
A new paper proposes rethinking supervision targets for multimodal LLM unlearning, showing that removing less private information can reduce distribution shift while addressing privacy risks from memorized sensitive data.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.CL
Rethinking Supervision Targets for Multimodal LLM Unlearning: Less Private Information, Reduced Distribution Shift