conftrace_

capability degradation

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

harmful instruction (1) harmful instruction defense (1) model defense (1) large language model (1) composed image retrieval (1) malicious fine-tuning (1) model capability (1) large language model alignment (1) continual training (1) fine-tuning attack (1)

Papers

ReCALL: Recalibrating Capability Degradation for MLLM-based Composed Image Retrieval CVPR 2026 SDD: Self-Degraded Defense against Malicious Fine-tuning ACL 2025