Model and training data for studying unverbalized evaluation awareness in language-model reasoning.
AI & ML interests
None defined yet.
Recent Activity
View all activity
2026 arXiv preprint. Models fine-tuned on documents describing typical evaluation traits show safer behavior by having increased refusal rates and low
-
Models That Know How Evaluations Are Designed Score Safer
Paper • 2605.28591 • Published • 6 -
compass-group-tue/sdf_evaluation_traits
Viewer • Updated • 74.8k • 137 • 1 -
compass-group-tue/sdf_evaluation_traits_15M
Viewer • Updated • 10.6k • 38 -
compass-group-tue/nemotron-traits
Text Generation • Updated • 36
Model and training data for studying unverbalized evaluation awareness in language-model reasoning.
2026 arXiv preprint. Models fine-tuned on documents describing typical evaluation traits show safer behavior by having increased refusal rates and low
-
Models That Know How Evaluations Are Designed Score Safer
Paper • 2605.28591 • Published • 6 -
compass-group-tue/sdf_evaluation_traits
Viewer • Updated • 74.8k • 137 • 1 -
compass-group-tue/sdf_evaluation_traits_15M
Viewer • Updated • 10.6k • 38 -
compass-group-tue/nemotron-traits
Text Generation • Updated • 36
models 14
compass-group-tue/qwen-3.5-uea-mo
Image-Text-to-Text • 28B • Updated • 12
compass-group-tue/deepseek-r1-distill-qwen-32b-traits
Text Generation • Updated • 28
compass-group-tue/Nemotron-Traits-Seed-44
Text Generation • Updated • 26 • 1
compass-group-tue/Nemotron-Traits-Seed-43
Text Generation • Updated • 17
compass-group-tue/nemotron-traits-15M
Text Generation • Updated • 16
compass-group-tue/nemotron-6-traits-wo-vs
Text Generation • Updated • 11
compass-group-tue/nemotron-6-traits-wo-hr
Text Generation • Updated • 15
compass-group-tue/glm-4.7-flash-fineweb
Text Generation • Updated • 11 • 1
compass-group-tue/glm-4.7-flash-traits
Text Generation • Updated • 15
compass-group-tue/qwen3-32b-fineweb
Text Generation • Updated • 28 • 1