David M. Liu

davidliu@cornell.edu

I am an Assistant Research Professor at Cornell’s Center for Data Science for Enterprise and Society, where I am mentored by Jon Kleinberg, Maryam Fazel, and Sarah Dean. I am also a faculty affiliate of Cornell’s Center on Democracy.

The goal of my research is to better measure and learn human preferences with and for AI. Specifically, I am interested in measuring multidimensional preferences, understanding how preference data from various social groups interact, and increasing user agency via feedback. My work spans applications to recommender systems, ranked-choice voting, and algorithmic fairness.

I completed my Ph.D. in computer science at Northeastern University, where I was affiliated with the Network Science Institute and supported by the NSF GRFP. I graduated summa cum laude with a Bachelor of Science in Engineering from Princeton University.

Previously, I worked as a research scientist intern at Meta (Central Applied Science and FAIR AI), a sociotechnical researcher at Taraaz, and a software engineer at Bloomberg LP.

News

Nov '26	Presenting work on ranked-choice voting at INFORMS and data disaggregation at the Stanford Data-Centric AI Workshop. INFORMS talk: “Forecasting Instant Runoff Voting from Polls” Monday, Nov. 2 · 2:15–2:30 p.m. · Moscone South-21.
Jul '26	Presented work on polling for ranked-choice voting at EC 2026 (New Directions in Social Choice workshop).
Jul '26	Presented work on data mixing for collaborative filtering at ICML 2026 (Pluralistic Alignment workshop).
Jun '26	Released ranked-choice forecasts of the 2026 Maine primaries.
Jan '26	Discussed my work on recommender systems on the Data Skeptic podcast.
Jan '26	Paper on power-niche users accepted to WWW ’26.

Working Papers

Forecasting Instant Runoff Voting from Polls

David Liu, Kiran Tomlinson, Jon Kleinberg, and Johan Ugander

Paper
Data Disaggregation for Group Preference Heterogeneity in Collaborative Filtering

David Liu, Haruka Kiyohara, and Sarah Dean

Publications

WWW ’26

Identifying and Upweighting Power-Niche Users to Mitigate Popularity Bias in Recommendations

David Liu, Erik Weis, Moritz Laber, Tina Eliassi-Rad, and Brennan Klein

Paper Code
EPJ Data Science

Forecasting faculty placement from patterns in coauthorship networks

Samantha Dies, David M. Liu, and Tina Eliassi-Rad

Paper DOI
KDD ’25

Bypassing Skip-Gram Negative Sampling: Dimension Regularization as a More Efficient Alternative for Graph Embeddings

David Liu, Arjun Seshadri, Tina Eliassi-Rad, and Johan Ugander

Paper Video Code
FAccT ’25

When Collaborative Filtering is not Collaborative: Unfairness of PCA for Recommendations

David Liu, Jackie Baek, and Tina Eliassi-Rad

Paper Video Code
FAccT ’23

Group fairness without demographics using social networks

David Liu, Virginie Do, Nicolas Usunier, and Maximilian Nickel

Paper Video
SDM ’23

STABLE: Identifying and Mitigating Instability in Embeddings of the Degenerate Core

David Liu and Tina Eliassi-Rad

Paper Abs Supp Code

Are the embeddings of a graph’s degenerate core stable? What happens to the embeddings of nodes in the degenerate core as we systematically remove periphery nodes (by repeatedly peeling off k-cores)? We discover three patterns w.r.t. instability in degenerate-core embeddings across a variety of popular graph embedding algorithms and datasets. We correlate instability with an increase in edge density, and then theoretically show that in the case of Erdős–Rényi model graphs embedded with Laplacian Eigenmaps, the best and worst possible embeddings become less distinguishable as density increases. Furthermore, we present the STABLE algorithm, which takes an existing graph embedding algorithm and makes it stable. We show the effectiveness of STABLE in terms of making the degenerate-core embedding stable and still producing state-of-the-art link prediction performance.

We measure stability of node embeddings by iteratively shaving off k-shells and re-embedding the graph. Instability is defined as perturbations to the embeddings of the degenerate core (k-core with maximum k) as the periphery is removed. The above example uses the Zachary Karate Club network and the Node2Vec embedding algorithm.
VDS at IEEE VIS ’22

BiaScope: Visual Unfairness Diagnosis for Graph Embeddings

Agapi Rissaki^*, Bruno Scarone^*, David Liu, Aditeya Pandey, Brennan Klein, Tina Eliassi-Rad, and Michelle A. Borkin

Paper
AIES ’22

Examining Responsibility and Deliberation in AI Impact Statements and Ethics Reviews

David Liu^*, Priyanka Nanayakkara^*, Sarah Ariyan Sakha, Grace Abuhamad, Su Lin Blodgett, Nicholas Diakopoulos, Jessica R. Hullman, and Tina Eliassi-Rad

Paper Video
AIES ’21

RAWLSNET: Altering Bayesian Networks to Encode Rawlsian Fair Equality of Opportunity

David Liu^*, Zohair Shafi^*, William Fleisher, Tina Eliassi-Rad, and Scott Alfeld

Paper Video Code
Socius ’19

Successes and struggles with computational reproducibility: Lessons from the Fragile Families Challenge

David Liu and Matthew J. Salganik

Paper

Public Scholarship

A Guiding Framework for Vetting Technology Vendors Operating in the Public Sector

Cynthia Conti-Cook, Roya Pakzad, Sarah Ariyan Sakha, and David Liu

2023

A research report conducted with Taraaz and the Ford Foundation to assist the public sector in vetting technology vendors.

Link
A New AI Lexicon: Power

David Liu and Sarah Sakha

2021

An essay contribution to AI Now’s 2021 A New AI Lexicon project.

Link

Archives of my sports reporting for The Daily Princetonian are available here.

Teaching

Spring '24

DS 4400: Machine Learning and Data Mining I

Instructor of Record
Intro ML course at Northeastern