-
The Allen Institute for Artificial Intelligence (AI2)
- https://www.krichardson.me/
Highlights
- Pro
Stars
Materials for course on language model programming (LMProgramming)
Source code and utilities for the Genesys distributed language model architecture discovery system.
[EMNLP 2025 Demo] TinyScientist: A Lightweight Framework for Building Research Agents
Extension to the built-in ast module. Finds comments in source code and adds them to the parsed tree.
Exact inference for discrete probabilistic programs. (Research code, more documentation and ergonomics to come)
PyTorch building blocks for the OLMo ecosystem
The Gradient of Algebraic Model Counting (AAAI2025)
Learning to route instances for Human vs AI Feedback (ACL Main '25)
Proof-of-concept formal verification using Hoare logic
a python framework to build, learn and reason about probabilistic circuits and tensor networks
Guardrails for secure and robust agent development
Awesome deliberative prompting: How to ask LLMs to produce reliable reasoning and make reason-responsive decisions.
A framework for evaluating the effectiveness of chain-of-thought reasoning in language models.
π€ Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
π« Industrial-strength Natural Language Processing (NLP) in Python
Original data for the text to code experiments of Richardson and Kuhn
Source code for our paper: "Put Your Money Where Your Mouth Is: Evaluating Strategic Planning and Execution of LLM Agents in an Auction Arena"
A language for constraint-guided and efficient LLM programming.
Course resources and notes for the ESSLLI 2023 course on neural symbolic methods.
Code for situation modeling, EMNLP 2022
PyPSDD porting to Python 3 + PyTorch equivalent tree construction.
π[ChatGPT4NLU] A Comparative Study on ChatGPT and Fine-tuned BERT