Configurable Preference Tuning with Rubric-Guided Synthetic Data
Paper
•
2506.11702
•
Published
•
2
Configurable Preference Tuning ⚙️📝
CPT uses rubric-guided synthetic data and DPO to enable LLMs to dynamically adjust behavior (e.g., writing style) at inference with system prompts
-
-
vicgalle/creative-rubrics-preferences
Viewer • Updated • 900 • 135 • 2 -
vicgalle/creative-rubrics
Viewer • Updated • 1.33k • 135 • 3 -
vicgalle/configurable-preference-mistral-nemo-12b
Text Generation • Updated -
vicgalle/configurable-preference-rocinante-12b
Text Generation • Updated • 19 -
vicgalle/configurable-preference-phi4
Text Generation • Updated -
vicgalle/configurable-preference-qwen3-4b
Text Generation • Updated