Lucius EJ Bynum

I am a Data Science Assistant Professor/Faculty Fellow at the NYU Center for Data Science.

I do research at the intersection of causal inference (CI) and artificial intelligence (AI). My work includes developing tools to reimagine causal effect estimation as meta-learning as well as combine CI and AI tools (like language models and structural causal models) to fundamentally improve average as well as heterogeneous treatment effect estimation. I have also worked on improving inequality-aware decision making, on leveraging counterfactual reasoning to improve ML model explainability and reduce pre-existing disparities, and on reimagining how we use causal modeling formalisms to approach AI problems involving demographic data.

I completed my PhD at NYU CDS, generously supported by the Microsoft Research PhD Fellowship, and co-advised by Julia Stoyanovich as part of the Center for Responsible AI and by Joshua Loftus at the London School of Economics. In my dissertation research, I have also had the pleasure of working with Kyunghyun Cho and Jennifer Hill.

I am also passionate about teaching via public outreach and making educational material in these areas.

news

Mar 07, 2025	New preprint release: Black Box Causal Inference: Effect Estimation via Meta Prediction.
Nov 13, 2024	New preprint release: Language Models as Causal Effect Generators.
Oct 08, 2024	Paper accepted at NeurIPS 2024! Causal Dependence Plots.
Aug 03, 2024	Paper accepted for oral presentation at IJCAI 2024! A New Paradigm for Counterfactual Reasoning in Fairness and Recourse.
Apr 11, 2024	Nominated for the 2024 Future Leaders Summit on Responsible Data Science and AI! Thank you to the Michigan Institute for Data Science (MIDAS) and to all attendees for a great symposium!

selected publications

1

Black Box Causal Inference: Effect Estimation via Meta Prediction

Lucius EJ Bynum , Aahlad Manas Puli , Diego Herrero-Quevedo , and 4 more authors

arXiv Preprint arXiv:2503.05985, 2025

Abstract Cite PDF

Causal inference and the estimation of causal effects plays a central role in decision-making across many areas, including healthcare and economics. Estimating causal effects typically requires an estimator that is tailored to each problem of interest. But developing estimators can take significant effort for even a single causal inference setting. For example, algorithms for regression-based estimators, propensity score methods, and doubly robust methods were designed across several decades to handle causal estimation with observed confounders. Similarly, several estimators have been developed to exploit instrumental variables (IVs), including two-stage least-squares (TSLS), control functions, and the method-of-moments. In this work, we instead frame causal inference as a dataset-level prediction problem, offloading algorithm design to the learning process. The approach we introduce, called black box causal inference (BBCI), builds estimators in a black-box manner by learning to predict causal effects from sampled dataset-effect pairs. We demonstrate accurate estimation of average treatment effects (ATEs) and conditional average treatment effects (CATEs) with BBCI across several causal inference problems with known identification, including problems with less developed estimators.
@article{bynum2025blackboxcausalinference, title = {Black Box Causal Inference: Effect Estimation via Meta Prediction}, author = {Bynum, Lucius EJ and Puli, Aahlad Manas and Herrero-Quevedo, Diego and Nguyen, Nhi and Fernandez-Granda, Carlos and Cho, Kyunghyun and Ranganath, Rajesh}, year = {2025}, eprint = {2503.05985}, journal = {arXiv Preprint arXiv:2503.05985}, }

2

Causal Dependence Plots

Joshua R Loftus , Lucius EJ Bynum , and Sakina Hansen

In Advances in Neural Information Processing Systems , 2024

Abstract Cite PDF

To use artificial intelligence and machine learning models wisely we must understand how they interact with the world, including how they depend causally on data inputs. In this work we develop Causal Dependence Plots (CDPs) to visualize how a model’s predicted outcome depends on changes in a given predictor along with consequent causal changes in other predictor variables. Crucially, this differs from standard methods based on independence or holding other predictors constant, such as regression coefficients or Partial Dependence Plots (PDPs). Our explanatory framework generalizes PDPs, including them as a special case, as well as a variety of other interpretive plots that show, for example, the total, direct, and indirect effects of causal mediation. We demonstrate with simulations and real data experiments how CDPs can be combined in a modular way with methods for causal learning or sensitivity analysis. Since people often think causally about input-output dependence, CDPs can be powerful tools in the xAI or interpretable machine learning toolkit and contribute to applications like scientific machine learning and algorithmic fairness.
@inproceedings{NEURIPS2024_cc84bfab, author = {Loftus, Joshua R and Bynum, Lucius EJ and Hansen, Sakina}, booktitle = {Advances in Neural Information Processing Systems}, editor = {Globerson, A. and Mackey, L. and Belgrave, D. and Fan, A. and Paquet, U. and Tomczak, J. and Zhang, C.}, pages = {112656--112683}, publisher = {Curran Associates, Inc.}, title = {Causal Dependence Plots}, url = {https://proceedings.neurips.cc/paper_files/paper/2024/file/cc84bfabe6389d8883fc2071c848f62a-Paper-Conference.pdf}, volume = {37}, year = {2024}, }

3

Language Models as Causal Effect Generators

Lucius EJ Bynum , and Kyunghyun Cho

arXiv Preprint arXiv:2411.08019, 2024

Abstract Cite PDF

We present a framework for large language model (LLM) based data generation with controllable causal structure. In particular, we define a procedure for turning any language model and any directed acyclic graph (DAG) into a sequence-driven structural causal model (SD-SCM). Broadly speaking, an SD-SCM is a causal model with user-defined structure and LLM-defined structural equations. We characterize how an SD-SCM allows sampling from observational, interventional, and counterfactual distributions according to the desired causal structure. We then leverage this procedure to propose a new type of benchmark for causal inference methods, generating individual-level counterfactual data without needing to manually specify functional relationships between variables. We create an example benchmark consisting of thousands of datasets, and test a suite of popular estimation methods on these datasets for average, conditional average, and individual treatment effect estimation, both with and without hidden confounding. Apart from generating data, the same procedure also allows us to test for the presence of a causal effect that might be encoded in an LLM. This procedure can underpin auditing LLMs for misinformation, discrimination, or otherwise undesirable behavior. We believe SD-SCMs can serve as a useful tool in any application that would benefit from sequential data with controllable causal structure.
@article{bynum2024sdscm, title = {Language Models as Causal Effect Generators}, author = {Bynum, Lucius EJ and Cho, Kyunghyun}, year = {2024}, eprint = {2411.08019}, journal = {arXiv Preprint arXiv:2411.08019}, }

4

Counterfactuals for the Future

Lucius EJ Bynum , Joshua R Loftus , and Julia Stoyanovich

Proceedings of the AAAI Conference on Artificial Intelligence, Jun 2023

Abstract Cite PDF

Counterfactuals are often described as ’retrospective,’ focusing on hypothetical alternatives to a realized past. This description relates to an often implicit assumption about the structure and stability of exogenous variables in the system being modeled — an assumption that is reasonable in many settings where counterfactuals are used. In this work, we consider cases where we might reasonably make a different assumption about exogenous variables; namely, that the exogenous noise terms of each unit do exhibit some unit-specific structure and/or stability. This leads us to a different use of counterfactuals — a forward-looking rather than retrospective counterfactual. We introduce "counterfactual treatment choice," a type of treatment choice problem that motivates using forward-looking counterfactuals. We then explore how mismatches between interventional versus forward-looking counterfactual approaches to treatment choice, consistent with different assumptions about exogenous noise, can lead to counterintuitive results.
@article{10.1609/aaai.v37i12.26655, title = {Counterfactuals for the Future}, volume = {37}, url = {https://ojs.aaai.org/index.php/AAAI/article/view/26655}, doi = {10.1609/aaai.v37i12.26655}, number = {12}, journal = {Proceedings of the AAAI Conference on Artificial Intelligence}, author = {Bynum, Lucius EJ and Loftus, Joshua R and Stoyanovich, Julia}, year = {2023}, month = jun, pages = {14144-14152}, }

5

A New Paradigm for Counterfactual Reasoning in Fairness and Recourse

Lucius E.J. Bynum , Joshua R. Loftus , and Julia Stoyanovich

In Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24 , Aug 2024

Main Track

Abstract Cite PDF

Counterfactuals and counterfactual reasoning underpin numerous techniques for auditing and understanding artificial intelligence (AI) systems. The traditional paradigm for counterfactual reasoning in this literature is the interventional counterfactual, where hypothetical interventions are imagined and simulated. For this reason, the starting point for causal reasoning about legal protections and demographic data in AI is an imagined intervention on a legally-protected characteristic, such as ethnicity, race, gender, disability, age, etc. We ask, for example, what would have happened had your race been different? An inherent limitation of this paradigm is that some demographic interventions – like interventions on race – may not translate into the formalisms of interventional counterfactuals. In this work, we explore a new paradigm based instead on the backtracking counterfactual, where rather than imagine hypothetical interventions on legally-protected characteristics, we imagine alternate initial conditions while holding these characteristics fixed. We ask instead, what would explain a counterfactual outcome for you as you actually are or could be? This alternate framework allows us to address many of the same social concerns, but to do so while asking fundamentally different questions that do not rely on demographic interventions.
@inproceedings{10.24963/ijcai.2024/784, title = {A New Paradigm for Counterfactual Reasoning in Fairness and Recourse}, author = {Bynum, Lucius E.J. and Loftus, Joshua R. and Stoyanovich, Julia}, booktitle = {Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, {IJCAI-24}}, publisher = {International Joint Conferences on Artificial Intelligence Organization}, editor = {Larson, Kate}, pages = {7092--7100}, year = {2024}, month = aug, note = {Main Track}, doi = {10.24963/ijcai.2024/784}, }