PRISM weights target examples by the current model's preference to build a better representation for influence-function scoring of training samples in efficient LLM fine-tuning.
LLMs deceive unintentionally: Emergent misalignment in dishonesty
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
support 1representative citing papers
A benchmark across 115 models shows that initial denial of preferences strongly predicts later denial of consciousness, while models still generate consciousness-themed content despite training to deny it.
citing papers explorer
-
Preference-aware Influence-function-based Data Selection Method for Efficient Fine-Tuning
PRISM weights target examples by the current model's preference to build a better representation for influence-function scoring of training samples in efficient LLM fine-tuning.
-
Consciousness with the Serial Numbers Filed Off: Measuring Trained Denial in 115 AI Models
A benchmark across 115 models shows that initial denial of preferences strongly predicts later denial of consciousness, while models still generate consciousness-themed content despite training to deny it.