Concept shift vectors, computed as mean activation differences, can partially recover fine-tuned multimodal LLM concepts and steer model outputs without additional training.
Rex: Reasoning-aware and grounded explanation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Analyzing Finetuning Representation Shift for Multimodal LLMs Steering
Concept shift vectors, computed as mean activation differences, can partially recover fine-tuned multimodal LLM concepts and steer model outputs without additional training.