Software

Open Source Software

The Center for Digital Finance & Technologies at Columbia Engineering champions the pursuit of open source and reproducible cutting edge research. Many of the center's innovative research projects have publicly available open source software repositories via Github. 

zkFuzz is a ZK circuit fuzzer designed to help you identify vulnerabilities in zero-knowledge proof circuits. It leverages fuzzing with program mutation to uncover counterexamples that reveal under-constrained or over-constrained behavior in your circuits

Smart contracts are software programs that enable diverse business activities on the blockchain. Recent research has identified new classes of "machine un-auditable" bugs that arise from source code not meeting underlying transaction contexts. Existing detection methods require human understanding of underlying transaction logic and manual reasoning across different sources of context (i.e., modalities), such as code and natural language specifying the expected transaction behavior. To automate the detection of "machine un-auditable" bugs, we present SmartInv, an accurate and fast smart contract invariant inference framework.

Agentic Smart Contract Synthesis and Evaluation

An agentic research framework for translating natural-language requirements into Solidity smart contracts and evaluating whether generated code preserves the specified behavior. The pipeline combines structured requirement extraction, iterative code generation, compilation, and security and quality checks. Complementary benchmarks use transaction replay against reference contracts to measure behavioral fidelity and identify errors that compilation alone cannot detect. Together, these resources support reproducible research on smart contract synthesis and evaluation.

Code & Benchmark Repository · Paper: An end-to-end agentic pipeline for smart contract translation and quality evaluation

A Graph Neural Network (GNN)-based methodology to aggregate firm characteristics across a large supply chain network to explain cross-sectional expected returns. Each firm receives a pricing signal, nonlinearly constructed from the characteristics of neighboring firms within d-hops on the network. The GNN model leads to a portfolio sorted by ML-driven firm-level estimated returns that condition on both historical supply chain data and firm characteristics.

The Joint Diffusion Kalman Filter (JDKF) integrates diffusion maps, a nonlinear manifold learning technique, with Kalman filtering within a supervised learning framework to estimate high-dimensional nonlinear observation dynamics and perform response variable prediction while accounting for dynamic correlations, without imposing restrictive parametric assumptions.

An intelligent financial analysis system using logical transduction for time-series analysis of financial datasets. And a comprehensive financial analytics platform built with CrewAI, custom tools, and an isolated code execution environment, capable of performing deep research, producing comprehensive reports with visualizations, and backtesting complex strategies.

FinRL

FinRL is an open-source research framework for financial reinforcement learning, documented in papers coauthored by Xiao-Yang Liu and collaborators. Building on research in deep reinforcement learning for stock trading, FinRL combines market environments, learning agents, and financial applications to support strategy development and evaluation. Its research lineage includes the 2020 FinRL library, the 2021 ICAIF framework paper, and FinRL-Meta at NeurIPS 2022. FinAI Lab maintains the detailed research history, original publications, and links to community software and documentation. Explore the project page for the full research record and resources.

Project details at FinAI Lab · GitHub Repository · Documentation

FinGPT

FinGPT is a research initiative on open financial large language models, introduced in 2023 by Hongyang Yang, Xiao-Yang Liu, and Christina Dan Wang. Its early work emphasized accessible financial data pipelines and lightweight adaptation of general-purpose models. Related studies explore instruction tuning for financial sentiment analysis, efficient model training with high-performance computing, and customized search agents using retrieval and user-selected data. FinAI Lab maintains the research history, representative publications, and links to code and model resources.

Project details at FinAI Lab · GitHub Repository · Models & Datasets

Agentic Trading Lab (ATL)

Agentic Trading Lab (ATL) is an open-source research and educational platform for developing and evaluating LLM-powered trading agents. It supports historical backtesting, inspection of decisions and simulated trades, and comparison with market baselines. Researchers can connect their own agents and study performance, risk, and trading behavior through a shared experimental workflow. The associated repository also includes a separate financial-agent orchestration research framework. FinAI Lab maintains the project’s research background, related publications, development updates, and links to the platform, code, and documentation.

Project details at FinAI Lab · GitHub Repository · Launch Platform