enochlev/llm-grpo-toddler-small-11
Text Generation • 0.4B • Updated
• 1
ToddlerLM
These are the models used for the paper: https://aclanthology.org/2025.sigdial-1.30/
Note Fine-tuned Smol small model.
enochlev/llm-toddler-30
Text Generation • 0.2B • Updated • 2 • 1Note pre-trained on CHILDES; best reported model in paper
enochlev/childish_behavior_model
Text Classification • 0.1B • UpdatedNote used for training; reward model for determining age of generated outputs from toddler model
enochlev/child_coherence_model
Text Classification • 0.1B • UpdatedNote used for training; reward model for determining coherence of generated outputs from toddler model
enochlev/caregiver-question-clarity
Text Classification • 0.1B • UpdatedNote model used to filter data