Jason Poulos

AssetOpsBench. AssetOpsBench - Industry 4.0: A unified benchmark and framework for building, orchestrating, and evaluating domain-specific AI agents for Industry 4.0 asset operations and maintenance, with 460+ scenarios, 5 specialist agents (IoT, FMSR, TSFM, Work Order,...), and multi-agent orchestration blueprints (MetaAgent, AgentHive) over MCP.

terminal-bench-3. Measuring agents' ability to get work done on a computer

endless-terminals. Python

FEC. A general discussion forum for FEC.gov. This is the best place to submit general feedback.

nanobanana. TypeScript

viberank. 🏆 The AI coding usage leaderboard — Claude Code, Codex, Gemini CLI & more. Real costs and tokens from ccusage data. Submit with: npx viberank-cli

political-neutrality-eval. This repo contains detailed implementation information about Anthropic's paired prompts approach for evaluating political neutrality.

terminal-bench-2. Shell

harbor. Framework for evaluating and improving agents

content-aware-inpainting-v2.

reasonIF. ReasonIF: Large Language Models Fail to Follow Instructions During Reasoning

arxiv-latex-cleaner. arXiv LaTeX Cleaner: Easily clean the LaTeX code of your paper to submit to arXiv

music-arena. Python

DeepSeek-OCR. Contexts Optical Compression

nanochat. The best ChatGPT that $100 can buy.

generative-simulations. generative-simulations

OpenTSLM. OpenTSLM: Time-Series Language Models for Reasoning over Multivariate Medical Text- and Time-Series Data

Executable_Counterfactuals.

tinker-cookbook. Post-training with Tinker

SkyRL. SkyRL: A Modular Full-stack RL Library for LLMs

Sci2Pol. Sci2Pol: Evaluating and Fine-tuning LLMs on Scientific-to-Policy Brief Generation

Paper2Agent. Paper2Agent is a multi-agent AI system that automatically transforms research papers into interactive AI agents with minimal human input.

multi-agent-coding-system. Reached #13 on Stanford's Terminal Bench leaderboard. Orchestrator, explorer & coder agents working together with intelligent context sharing.

dclm. DataComp for Language Models

d2. D2 is a modern diagram scripting language that turns text to diagrams.

math-game-cli. Math game in the cli

SAITS. The official PyTorch implementation of the paper "SAITS: Self-Attention-based Imputation for Time Series". A fast and state-of-the-art (SOTA) deep-learning neural network model for efficient time-series imputation (impute multivariate incomplete time series containing NaN missing data/values with machine learning). https://arxiv.org/abs/2202.08516

PyPOTS. A Python toolkit/library for reality-centric machine/deep learning & data mining on partially-observed time series, with 50+ SOTA neural network models for scientific analysis tasks (imputation, classification, clustering, forecasting, anomaly detection, cleaning) on incomplete industrial irregularly-sampled multivariate TS with NaN missing values

terminal-bench-leaderboard. A home for run logs for terminal-bench

Self-Evolving-Agents.

terminal-bench. A benchmark for LLMs on complicated tasks in the terminal

TextArena. A Collection of Competitive Text-Based Games for Language Model Evaluation and Reinforcement Learning

lord_rust. Legend of the Red Dragon (Rust Edition)

textplayer. Batch running text-based games in Frotz

textquests. ZAP

BattleAgent. Python

WarAgent. WarAgent: LLM-based Multi-Agent Simulation of World Wars

heurigym. Agentic Benchmark for LLM-Crafted Heuristics in Combinatorial Optimization (ICLR'26)

HistAgent. Python

terminal-bench-rl. GRPO training code which scales to 32xH100s for long horizon terminal/coding tasks. Base agent is now the top Qwen3 agent on Stanford's TerminalBench leaderboard.

cfcausal. R package cfcausal

tabarena. A Living Benchmark for Machine Learning on Tabular Data

gemini-cli. An open-source AI agent that brings the power of Gemini directly into your terminal.

CL4R1T4S. LEAKED SYSTEM PROMPTS FOR CHATGPT, CLAUDE, GEMINI, GROK, PERPLEXITY, CURSOR, LOVABLE, REPLIT, AND MORE! - AI SYSTEMS TRANSPARENCY FOR ALL! 👐

codex. Lightweight coding agent that runs in your terminal

grok-prompts. Prompts for our Grok chat assistant and the `@grok` bot on X.

coupled-llm-evaluation. Code for "Evaluation of Large Language Models via Coupled Token Generation", AISTATS 2026.

simple-evals. Python

PODS. PODS: Potential Outcomes for Deferring Systems

matched_CFR. Python

ehrshot-benchmark. A benchmark for few-shot evaluation of foundation models for electronic health records (EHRs)

M3-IMPUTE. Code for the paper "$M^3$-Impute: Mask-guided Representation Learning for Missing Value Imputation"

sequence-driven-scms. Code for "Language Models as Causal Effect Generators" implementing sequence-driven structural causal models (SD-SCMs)

ScholarCopilot. ScholarCopilot: Training Large Language Models for Academic Writing with Accurate Citations [COLM 2025]

context-is-key-forecasting. Context is Key: A Benchmark for Forecasting with Essential Textual Information

llm_processes. Python

pywhyllm. Experimental library integrating LLM capabilities to support causal analyses

CausalNeuralSurvivalClustering. Implementation of "Identifying treatment response subgroups in observational time-to-event data" Jeanselme et al.

transformer-lm-gan. Explorations into adversarial losses on top of autoregressive loss for language modeling

improving-transformers-world-model-for-rl. Implementation of the new SOTA for model based RL, from the paper "Improving Transformer World Models for Data-Efficient RL", in Pytorch

chirho. An experimental language for causal reasoning

mostlyai. Synthetic Data SDK ✨

DeepSeek-R1.

courses. Anthropic's educational courses

curator. Synthetic data curation for post-training and structured data extraction

ellmer. Call LLM APIs from R

MEDS-DEV. The MEDS Decentralized Extensible Validation (MEDS-DEV) Benchmark: Establishing Reproducibility and Comparability in ML for Health

MetaGPT. 🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

TabCSDI. A code for the NeurIPS 2022 Table Representation Learning Workshop paper: "Diffusion models for missing value imputation in tabular data"

spline-based-transformer. Implementation of the proposed Spline-Based Transformer from Disney Research

chronos-forecasting. Chronos: Pretrained Models for Time Series Forecasting

nanoGPT. The simplest, fastest repository for training/finetuning medium-sized GPTs.

ethos-paper. Jupyter Notebook

ppi_py. A package for statistically rigorous scientific discovery using machine learning. Implements prediction-powered inference.

simglucose. A Type-1 Diabetes simulator implemented in Python for Reinforcement Learning purpose

MEDS_transforms. A simple set of MEDS polars-based ETL and transformation functions

meds_reader. A high-performance database and Python reader interface for MEDS datasets

meds-torch. Python

synthetic_prophet. Repository for the synthetic time series generation for Neural Prophet

AI-Scientist. The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑‍🔬

ACES. [ICLR 2025] ACES: Automatic Cohort Extraction System for Event-Streams

LLM101n. LLM101n: Let's build a Storyteller

Foresight. Deep Generative Modelling of Patient Timelines using Electronic Health Records

TransformEHR. Python

pygformula. The pygformula implements the parametric g-formula in Python. The parametric g-formula (Robins, 1986) uses longitudinal data with time-varying treatments and confounders to estimate the risk or mean of an outcome under hypothetical treatment strategies specified by the user.

MIMICIV_FMs_public. Sample end-to-end pipeline over MIMIC-IV demonstrating the Event Stream GPT code base.

bitsandbytes. Accessible large language models via k-bit quantization for PyTorch.

DeepSpeed. DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

flash-attention. Fast and memory-efficient exact attention

electra-pytorch. A simple and working implementation of Electra, the fastest way to pretrain language models from scratch, in Pytorch

bertviz. BertViz: Visualize Attention in Transformer Models

UniTS. A unified multi-task time series model.

grok-1. Grok open release

pytorch_tabular. A unified framework for Deep Learning Models on tabular data

saint. The official PyTorch implementation of recent paper - SAINT: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pre-Training

drotr. Doubly-robust methods for creating optimal treatment rules

refined-dpo. Refined Direct Preference Optimization with Synthetic Data for Behavioral Alignment of LLMs

rtdl. Research on Tabular Deep Learning: Papers & Packages

tab-transformer-pytorch. Implementation of TabTransformer, attention network for tabular data, in Pytorch

pyelixhauser. Python package to detect clinical morbodity in ICD10 and ICD9 codes

multivariate-attention-tcn. Jupyter Notebook

ViTST. [NeurIPS 2023] The official repo for the paper: "Time Series as Images: Vision Transformer for Irregularly Sampled Time Series"."

llm-course. Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

Awesome-LLM-Safety. A curated list of safety-related papers, articles, and resources focused on Large Language Models (LLMs). This repository aims to provide researchers, practitioners, and enthusiasts with insights into the safety implications, challenges, and advancements surrounding these powerful models.

be_great. A novel approach for synthesizing tabular data using pretrained large language models

TabSurvey. Experiments on Tabular Data Models

rpdrtools. Library of utility functions for handling RPDR (Research Patient Data Registry) data.

parseRPDR. Parse and Manipulate Research Patient Data Registry ('RPDR') Text Queries

GenePT. Jupyter Notebook

TimeSeries. Implementation of deep learning models for time series in PyTorch.

TESAR-CDE. Code for the paper "Accounting For Informative Sampling When Learning to Forecast Treatment Outcomes Over Time" (ICML 2023)

iTransformer. Unofficial implementation of iTransformer - SOTA Time Series Forecasting using Attention networks, out of Tsinghua / Ant group

SurvCI. SurvCI & SurvCI-Info:Counterfactual Inference using Balanced Representations for Parametric Deep Survival Analysis

ollama. Get up and running with Kimi-K2.6, GLM-5.1, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

tsgm. Generation and evaluation of synthetic time series datasets (also, augmentations, visualizations, a collection of popular datasets) NeurIPS'24

EHRMiss. An R package for Bayesian growth mixture modeling in electronic health records accounting for different missing data mechanisms for the visit process and the response process given a clinic visit

TimeSynth. A Multipurpose Library for Synthetic Time Series Generation in Python

medalign. MedAlign is a clinician-generated dataset for instruction following with electronic medical records.

Asclepius. Official Codes for "Publicly Shareable Clinical Large Language Model Built on Synthetic Clinical Notes"

Crossformer. Official implementation of our ICLR 2023 paper "Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting"

corr2cause. Data and code for the Corr2Cause paper (ICLR 2024)

llm-attacks. Universal and Transferable Attacks on Aligned Language Models

multivariate-time-series-data.

LSTNet. Python

conformal-time-series. Conformal prediction for time-series applications.

Transformer_Time_Series. Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting (NeurIPS 2019)

MachineControl. An R package to make causal inference from a panel in the absence of controls

TimeSeries_Seq2Seq. This repo aims to be a useful collection of notebooks/code for understanding and implementing seq2seq neural networks for time series forecasting. Networks are constructed with keras/tensorflow.

quivr. Opiniated RAG for integrating GenAI in your apps 🧠 Focus on your product rather than the RAG. Easy integration in existing products with customisation! Any LLM: GPT4, Groq, Llama. Any Vectorstore: PGVector, Faiss. Any Files. Anyway you want.

Awesome-LLM-Uncertainty-Reliability-Robustness. Awesome-LLM-Robustness: a curated list of Uncertainty, Reliability and Robustness in Large Language Models

Awesome-LLM4RS-Papers. Large Language Model-enhanced Recommender System Papers

Awesome-LLM. Awesome-LLM: a curated list of Large Language Model

imtp. :package: Incremental propensity score interventions as described in Nonparametric Causal Effects Based on Incremental Propensity Score Interventions (Kennedy, 2019)

lcmmtp. Longitudinal Mediation with Modified Treatment Policies

EventStreamGPT. Dataset and modelling infrastructure for modelling "event streams": sequences of continuous time, multivariate events with complex internal dependencies.

Dynamic-DeepHit. Dynamic-DeepHit: A Deep Learning Approach for Dynamic Survival Analysis With Competing Risks Based on Longitudinal Data

DeepHit. DeepHit: A Deep Learning Approach to Survival Analysis with Competing Risks

openai-cookbook. Examples and guides for using the OpenAI API

llm-security. New ways of breaking app-integrated LLMs

do-we-need-attention. TeX

chatblade. A CLI Swiss Army Knife for ChatGPT

gpt4all. GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.

leaked-system-prompts. Collection of leaked system prompts

SyGNet. Synthetic data using Generative Adversarial Networks

DeepSurv. DeepSurv is a deep learning approach to survival analysis.

SuMo-net. Monotone neural networks for survival analysis

generative-ai. Materials for "Generative AI for R programmers" workshop

AI-Pokemon-Music. Using AI (Transformers) to make original/ recreate Pokémon music

rcps. Official codebase for "Distribution-Free, Risk-Controlling Prediction Sets"

RNN-Time-series-Anomaly-Detection. RNN based Time-series Anomaly detector model implemented in Pytorch.

ricu. 🏥 ICU data with R 🏥

survdr. Targeted MLE with doubly robust asymptotic distribution for estimation of survival probabilities from right censored data

CausalSurvival. R pkg for estimating counterfactual survival curves using state-of-the-art causal inference methods

unihtee. Tools for uncovering treatment effect modifiers in high-dimensional data.

lmtp-tutorial. Corresponding code guide to the tutorial paper "Introducing longitudinal modified treatment policies: a unified framework for studying complex exposures" (Hoffman et al., 2023)

adjustedCurves. An R-Package to estimate and plot confounder-adjusted survival curves (single event survival data) and confounder-adjusted cumulative incidence functions (data with competing risks) using various methods.

lmtp-simulations. Code used for the simulation study in the paper "Non-parametric casual effects based on longitudinal modified treatment policies."

causalgraphicalmodels. Causal Graphical Models in Python

temporai. TemporAI: ML-centric Toolkit for Medical Time Series

simMSM. R package that simulates data suitable for fitting Marginal Structural Model.

drnns-prediction. Deep Recurrent Neural Networks (RNNs) for Time-Series Prediction

ga-lottery. Replication code for "Land Lotteries, Long-term Wealth, and Political Selection"

patt-c. Replication code for "Estimating population average treatment effects from experiments with noncompliance"

amnesty. Replication code for "Amnesty Policy and Elite Persistence in the Postbellum South"

ok-lottery. Code and data for the paper "Gender gaps in wealth and entrepreneurship? Evidence from 1901 Oklahoma land lottery winners"

homesteads. Replication code for "State-Building through Public Land Disposal? An Application of Matrix Completion for Counterfactual Prediction"

MAPIE. A scikit-learn-compatible library for estimating prediction intervals and controlling risks, based on conformal predictions.

causica. Python

causal-ml. Must-read papers and resources related to causal inference and machine (deep) learning

rnns-causal. Replication code for "RNN-based counterfactual prediction, with an application to homestead policy and public schooling"

gptstudio. GPT RStudio addins that enable GPT assisted coding, writing & analysis

ChatReviewer. ChatReviewer: 使用ChatGPT分析论文优缺点，提出改进建议

multi-tmle. Code and simulated data accompanying the paper "Targeted learning in observational studies with multi-level treatments: An evaluation of antipsychotic drug treatment safety"

FreeLunchSemiParametrics. Jupyter Notebook

Causality-in-Trustworthy-Machine-Learning. The repository contains lists of papers on causality and how relevant techniques are being used to further enhance deep learning era computer vision solutions.

transportability-workshop. Transporting treatment effects from randomized trials to real-world populations, Karolinska Institutet, 16th Nov 21

ACIC23-competition. Data for and description of the ACIC 2023 data competition

intro_dgm. "Deep Generative Modeling": Introductory Examples

causalML-teaching. This repository consolidates my teaching material for "Causal Machine Learning".

mcm. Jupyter Notebook

DeepMed. R

rap-rank-reconstruction. Code for reproducing https://arxiv.org/abs/2211.03128

Variational-Lstm-Autoencoder. Lstm variational auto-encoder for time series anomaly detection and features extraction

medgan. Generative adversarial network for generating electronic health records.

CompCATE. Code to replicate the results in the AISTATS23 paper "Understanding the Impact of Competing Events on Heterogeneous Treatment Effect Estimation from Time-to-Event Data" (Curth & van der Schaar, 2023)

tableGAN. tableGAN is a synthetic data generation technique (Data Synthesis based on Generative Adversarial Networks paper) based on Generative Adversarial Network architecture (DCGAN).

time-series-transformers-review. A professionally curated list of awesome resources (paper, code, data, etc.) on transformers in time series.

info8010-deep-learning. Lectures for INFO8010 Deep Learning, ULiège

SharedForestPaper. Code for the shared forest paper.

doubleml-for-r. DoubleML - Double Machine Learning in R

chalk-talk. TeX

LTSF-Linear. [AAAI-23 Oral] Official implementation of the paper "Are Transformers Effective for Time Series Forecasting?"

CausalTransformer. Code for the paper "Causal Transformer for Estimating Counterfactual Outcomes"

audio-ai-timeline. A timeline of the latest AI models for audio generation, starting in 2023!

keras-cv. Industry-strength Computer Vision workflows with Keras

cqr. Conformalized Quantile Regression

survivalgan. SurvivalGAN: Generating Time-to-Event Data for Survival Analysis

prediction-powered-inference. A statistical toolkit for scientific discovery using machine learning

ents. R implementation of maximum entropy synthetic controls

representation-causal-public. Jupyter Notebook