📊 Results
We fine-tuned small LLMs like Llama-3.1-8B-Instruct on MuCE using CrPO and achieved significant improvements on LLM output creativity across all dimensions while maintaining high output quality
CrPO models beat SFT, vanilla DPO, and large models such as GPT-4o 🔥
We fine-tuned small LLMs like Llama-3.1-8B-Instruct on MuCE using CrPO and achieved significant improvements on LLM output creativity across all dimensions while maintaining high output quality
CrPO models beat SFT, vanilla DPO, and large models such as GPT-4o 🔥
September 22, 2025 at 1:43 PM
📊 Results
We fine-tuned small LLMs like Llama-3.1-8B-Instruct on MuCE using CrPO and achieved significant improvements on LLM output creativity across all dimensions while maintaining high output quality
CrPO models beat SFT, vanilla DPO, and large models such as GPT-4o 🔥
We fine-tuned small LLMs like Llama-3.1-8B-Instruct on MuCE using CrPO and achieved significant improvements on LLM output creativity across all dimensions while maintaining high output quality
CrPO models beat SFT, vanilla DPO, and large models such as GPT-4o 🔥
📃Multi-task Creativity Evaluation (MuCE)
To apply CrPO, we also collect a large-scale preference dataset consisting of more than 200K human responses and ratings for more than 30 creativity assessments, and use a subset of it to train and evaluate our models.
To apply CrPO, we also collect a large-scale preference dataset consisting of more than 200K human responses and ratings for more than 30 creativity assessments, and use a subset of it to train and evaluate our models.
September 22, 2025 at 1:43 PM
📃Multi-task Creativity Evaluation (MuCE)
To apply CrPO, we also collect a large-scale preference dataset consisting of more than 200K human responses and ratings for more than 30 creativity assessments, and use a subset of it to train and evaluate our models.
To apply CrPO, we also collect a large-scale preference dataset consisting of more than 200K human responses and ratings for more than 30 creativity assessments, and use a subset of it to train and evaluate our models.
🧠 How do we compute creativity scores?
Instead of treating creativity as a single concept, we break it down into its major dimensions and employ metrics for each that provide measurable signals aligning with key cognitive theories and enable practical optimization within LLMs.
Instead of treating creativity as a single concept, we break it down into its major dimensions and employ metrics for each that provide measurable signals aligning with key cognitive theories and enable practical optimization within LLMs.
September 22, 2025 at 1:43 PM
🧠 How do we compute creativity scores?
Instead of treating creativity as a single concept, we break it down into its major dimensions and employ metrics for each that provide measurable signals aligning with key cognitive theories and enable practical optimization within LLMs.
Instead of treating creativity as a single concept, we break it down into its major dimensions and employ metrics for each that provide measurable signals aligning with key cognitive theories and enable practical optimization within LLMs.
🔧 How does it work?
CrPO = Direct Preference Optimization (DPO) × a weighted mix of creativity scores (novelty, surprise, diversity, quality).
This modular objective enables us to optimize LLMs for different dimensions of creativity tailored to a given domain.
CrPO = Direct Preference Optimization (DPO) × a weighted mix of creativity scores (novelty, surprise, diversity, quality).
This modular objective enables us to optimize LLMs for different dimensions of creativity tailored to a given domain.
September 22, 2025 at 1:43 PM
🔧 How does it work?
CrPO = Direct Preference Optimization (DPO) × a weighted mix of creativity scores (novelty, surprise, diversity, quality).
This modular objective enables us to optimize LLMs for different dimensions of creativity tailored to a given domain.
CrPO = Direct Preference Optimization (DPO) × a weighted mix of creativity scores (novelty, surprise, diversity, quality).
This modular objective enables us to optimize LLMs for different dimensions of creativity tailored to a given domain.
💡Can we optimize LLMs to be more creative?
Introducing Creative Preference Optimization (CrPO) and MuCE (Multi-task Creativity Evaluation Dataset).
Result: More novel, diverse, surprising text—without losing quality!
📝 Appearing at #EMNLP2025
Introducing Creative Preference Optimization (CrPO) and MuCE (Multi-task Creativity Evaluation Dataset).
Result: More novel, diverse, surprising text—without losing quality!
📝 Appearing at #EMNLP2025
September 22, 2025 at 1:43 PM
💡Can we optimize LLMs to be more creative?
Introducing Creative Preference Optimization (CrPO) and MuCE (Multi-task Creativity Evaluation Dataset).
Result: More novel, diverse, surprising text—without losing quality!
📝 Appearing at #EMNLP2025
Introducing Creative Preference Optimization (CrPO) and MuCE (Multi-task Creativity Evaluation Dataset).
Result: More novel, diverse, surprising text—without losing quality!
📝 Appearing at #EMNLP2025
Our analysis shows that model performance is negatively correlated with the morphological complexity of the words (i.e. number of morphemes) while human performance is not systematically affected (results shown for GPT-4 below)
February 20, 2025 at 5:28 PM
Our analysis shows that model performance is negatively correlated with the morphological complexity of the words (i.e. number of morphemes) while human performance is not systematically affected (results shown for GPT-4 below)
We find that all models struggle to compose new words and fail to consistently recognize the validity of all compositions, especially when applied to novel (i.e. out-of-distribution) word roots. Humans on the other hand ace both tasks and easily generalize to novel words.
February 20, 2025 at 5:28 PM
We find that all models struggle to compose new words and fail to consistently recognize the validity of all compositions, especially when applied to novel (i.e. out-of-distribution) word roots. Humans on the other hand ace both tasks and easily generalize to novel words.
We design two novel compositional probing tasks to measure morphological productivity (i.e. ability to produce novel well-formed combinations of morphemes) and systematicity (i.e. ability to systematically understand novel combinations)...
February 20, 2025 at 5:28 PM
We design two novel compositional probing tasks to measure morphological productivity (i.e. ability to produce novel well-formed combinations of morphemes) and systematicity (i.e. ability to systematically understand novel combinations)...