Melissa Roemmele

About

Melissa Roemmele works in AI engineering and ML & data science, based in San Francisco, CA. Research Scientist in Generative AI. Public projects on GitHub include Keras Rnn Notebooks (Python notebooks that demonstrate simple RNN models for NLP tasks using Keras), Narrative Scene Illustration (Code and prompts for the paper "LLMs Behind the Scenes: Enabling Narrative Scene Illustration") and Envisioning-Stories-with-Generative-AI (All materials related to the presentation "Envisioning Stories with Generative AI"), mostly written in Python, Jupyter and HTML.

Projects 6

• 

  Keras Rnn Notebooks

  Python notebooks that demonstrate simple RNN models for NLP tasks using Keras

  github.com
• github.com

  Narrative Scene Illustration

  Code and prompts for the paper "LLMs Behind the Scenes: Enabling Narrative Scene Illustration"

• 

  Envisioning-Stories-with-Generative-AI

  All materials related to the presentation "Envisioning Stories with Generative AI"

  github.com
• 

  Gen-COPA

  Using LLMs to generate items in the style of the Choice of Plausible Alternatives (COPA) evaluation benchmark

  github.com
• github.com

  Narrative Prediction

  Neural network models that predict "what happens next" in stories

• github.com

  Heider Simmel Recognition

  Code that uses a neural network to automatically assign anthropomorphic action labels (e.g. run, chase, fight) to animations of abstract moving shapes