Papers
See also Google Scholar See homepage for representative papers. "Lean" links refer to a Lean formalization of the paper's mathematical statements, using the EconCSLib project.
Working Papers
-
W8The Forest from the Trees: Lessons on Public-Sector Deployment from an Academic Collaboration with NYC Department of Parks and Recreation
-
W7What’s Wrong With Surveillance Pricing?California Law Review, 2027 (Pending official Editorial Board selection)
-
W2The Subjectivity of Monoculture
2026
-
C34Measuring the Impact of Urban Cycling Infrastructure with Large Scale E-Bike GPS DataEAAMO‘26Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO‘26)
-
C33Personalized Recommendations without Inducing Congestion: Mitigating Disparities in the NYC High School MatchEC‘26ACM Conference on Economics and Computation (EC‘26)EC Best Paper with Student Lead Author, 2026 (2 awards out of 1,111 submissions) INFORMS Doing Good with Good OR Student Paper Award, 2026 (Finalist; winners TBA) Plenary talk at IC2S2‘26 (top 2.3% of submissions)
-
J15Connecting Application Behavior to Undermatching in New York City School ChoiceNature Cities, Accepted
-
J14Combating Gerrymandering with Ranked Choice Voting: An Experimental Analysis of Multimember Districts in the United StatesOperations ResearchConference version published in ACM Conference on Economics and Computation (EC‘22) Cornell Chronicle, San Francisco Chronicle.
-
J13Dropping Standardized Testing for Admissions Trades Off Information and AccessManagement ScienceConference version published in ACM Conference on Fairness, Accountability, and Transparency (FAccT‘21); also appeared in EAAMO‘21 and in the 2021 NBER Decentralization Conference ACM EAAMO Student Paper Award, 2021 INFORMS DEI Best Student Paper Award, 2022 Yale Insights.
-
J12Improving flood detection with large-scale dashboard camera dataNature Communications
-
C32Position: Use Sparse Autoencoders to Discover UnknownsICML‘26 Position Paper TrackInternational Conference on Machine Learning (ICML‘26 Position Paper Track)
-
C31How Many Features Can a Language Model Store Under the Linear Representation Hypothesis?COLT‘26Conference on Learning Theory (COLT‘26)Spotlight Talk at ICML Workshop on Compositional Learning (3 of 210 submissions)
-
J11Simpler Than You Think: The Practical Dynamics of Ranked Choice VotingJournal of Computational Social Science
-
C30Urban Incident Prediction with Graph Neural Networks: Integrating Government Ratings and Crowdsourced ReportsAAAI‘26AAAI Conference on Artificial Intelligence (AAAI‘26)
-
C29Capacity Constraints Make Admissions Processes Less PredictableAAAI‘26AAAI Conference on Artificial Intelligence (AAAI‘26)Oral presentation.
2025
-
J10Inferring fine-grained migration patterns across the United StatesNature CommunicationsAssociation of American Geographers Best Student Paper Award, 2026 (Awardee: Gabriel Agostini) IPUMS Spatial Student Research Award, 2026 (Awardee: Gabriel Agostini) Cornell Chronicle.
-
J9"Shopping Around": An Experiment in Preferences and Incentives for Placing Long-term PatientsProceedings of the ACM on Human-Computer InteractionJournal Track for 28th ACM SIGCHI Conference on Computer-Supported Cooperative Work & Social Computing (CSCW‘25) Impact Recognition at CSCW ‘25 Cornell Chronicle.
-
O6Optimizing Library Usage and Browser Experience: Application to the New York Public Library
-
I1Heterogeneous participation and allocation skews: when is choice “worth it”?ACM SIGecom ExchangesInvited; lightly reviewed by editors
-
C28Correlated Errors in Large Language ModelsICML‘25International Conference on Machine Learning (ICML‘25)
-
C27Sparse Autoencoders for Hypothesis GenerationICML‘25International Conference on Machine Learning (ICML‘25)
-
C26Learning Disease Progression Models That Capture Health DisparitiesCHIL ‘25Conference on Health, Inference, and Learning (CHIL ‘25)Best Paper Award for Impact & Society at CHIL ‘25
-
C25Balancing Producer Fairness and Efficiency via Prior-Weighted Rating System DesignICWSM ‘25International AAAI Conference on Web and Social Media (ICWSM ‘25)
-
J8Faster Information for Effective Long-Term Discharge: A Field Study in Adult Foster CareProceedings of the ACM on Human-Computer InteractionJournal Track for 28th ACM SIGCHI Conference on Computer-Supported Cooperative Work & Social Computing (CSCW‘25). Also appeared in EAAMO‘23 Best Paper Award, and Diversity & Inclusion Recognition at CSCW ‘25 Cornell Chronicle.
-
C24A No Free Lunch Theorem for Human-AI CollaborationAAAI‘25AAAI Conference on Artificial Intelligence (AAAI‘25)
-
J7Addressing Discretization-Induced Bias in Demographic PredictionPNAS NexusConference version appeared in ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT 2024).
2024
-
C23Monoculture in Matching MarketsNeurIPS ‘24Neural Information Processing Systems (NeurIPS ‘24)
-
C22User-item fairness tradeoffs in recommendationsNeurIPS ‘24Neural Information Processing Systems (NeurIPS ‘24)
-
C21Ending Affirmative Action Harms Diversity Without Improving Academic MeritEAAMO ’24Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO ’24)
-
C20Wisdom and Foolishness of Noisy Matching MarketsEC‘24ACM Conference on Economics and Computation (EC‘24)
-
C19Equitable Congestion Pricing under the Markovian Traffic Model: An Application to BogotaEC‘24ACM Conference on Economics and Computation (EC‘24)
-
C18Topics, Authors, and Institutions in Large Language Model Research: Trends from 17K arXiv PapersNAACL‘24The North American Chapter of the Association for Computational Linguistics (NAACL‘24)
-
C17Domain constraints improve risk prediction when outcome data is missingICLR‘24International Conference on Learning Representations (ICLR‘24)
-
C16Reconciling the accuracy-diversity trade-off in recommendationsWWW‘24The ACM Web Conference (WWW‘24)Oral Presentation
-
C15A Bayesian Spatial Model to Correct Under-Reporting in Urban CrowdsourcingAAAI‘24AAAI Conference on Artificial Intelligence (AAAI‘24)Oral Presentation
-
C14Identifying and Addressing Disparities in Public Libraries with Bayesian Latent Variable ModelingAAAI‘24AAAI Conference on Artificial Intelligence (AAAI‘24)
-
O5Choosing the Right Weights: Balancing Value, Strategy, and Noise in Recommender Systems
2023
-
C13Supply-Side Equilibria in Recommender SystemsNeurIPS ‘23Neural Information Processing Systems (NeurIPS ‘23)
-
J6Quantifying Spatial Under-reporting Disparities in Resident CrowdsourcingNature Computational ScienceConference version published in ACM Conference on Economics and Computation (EC‘22), titled “Equity in Resident Crowdsourcing: Measuring Under-reporting without Ground Truth Data” INFORMS Junior Faculty Interest Group Paper Award (Finalist), 2023 Cornell Chronicle.
-
C12Interface Design to Mitigate Inflation in Recommender SystemsRecSys ’23 Short paperACM Conference on Recommender Systems (RecSys ’23 Short paper)
-
C11Coarse race data conceals disparities in clinical risk score performanceML4HC‘23Machine Learning for Healthcare (ML4HC) (ML4HC‘23)ML4H Best Findings Paper (Honorable Mention), 2023 Cornell Chronicle
2022
-
C10Fair ranking: a critical review, challenges, and future directionsFAccT‘22ACM Conference on Fairness, Accountability, and Transparency (FAccT‘22)This work was written as part of a distributed, student-led working group of Mechanism Design for Social Good
-
C9Trucks Don’t Mean Trump: Diagnosing Human Error in Image AnalysisFAccT‘22ACM Conference on Fairness, Accountability, and Transparency (FAccT‘22)
-
J5Driver Surge PricingManagement ScienceConference version published in EC‘20.
-
C8Strategic RankingAISTATS‘22International Conference on Artificial Intelligence and Statistics (AISTATS‘22)
2021
-
C7Test-optional Policies: Overcoming Strategic Behavior and Informational GapsEAAMO‘21AAAI/ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO‘21)
-
C6The Stereotyping Problem in Collaboratively Filtered Recommender SystemsEAAMO‘21AAAI/ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO‘21)
-
J4Designing Informative Rating Systems: Evidence from an Online Labor MarketManufacturing & Service Operations ManagementNew York Times, Stanford Engineering magazine. M&SOM student paper award (2nd place), 2020 Conference version published in EC‘20.
2020
-
J3Markets for Public Decision-makingSocial Choice and WelfareConference version published in WINE‘18.
-
T2Designing Marketplaces and Civic Engagement Platforms: Learning, Incentives, and PricingPhD Dissertation, Stanford University INFORMS George Dantzig Dissertation Award, 2020 ACM SIGecom dissertation award (Honorable mention), 2021
-
C5Fair Allocation through Selective Information AcquisitionAIES‘20AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES‘20)
2019
-
C4Who is in Your Top Three? Optimizing Learning in Elections with Many CandidatesHCOMP‘19AAAI Conference on Human Computation and Crowdsourcing (HCOMP‘19)
-
O4Deliberative Democracy with the Online Deliberation PlatformHCOMP‘19 DemoAAAI Conference on Human Computation and Crowdsourcing Demo Track (HCOMP‘19 Demo)
-
C3Analyzing Polarization in Social Media: Method and Application to Tweets on 21 Mass ShootingsNAACL‘19Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL‘19)
-
C2Designing Optimal Binary Rating SystemsAISTATS‘19International Conference on Artificial Intelligence and Statistics (AISTATS‘19)
-
J2Iterative Local Voting for Collective Decision-making in Continuous SpacesJournal of Artificial Intelligence Research (JAIR)Conference version published in WWW‘17.
2018
-
J1Word Embeddings Quantify 100 Years of Gender and Ethnic StereotypesProceedings of the National Academy of Sciences (PNAS)
-
O3Comparing Voting Methods for Budget Decisions on the ASSU Ballot
2015
-
C1Impact of Dual Slope Path Loss on User Association in HetNetsIEEE Globecom WorkshopIEEE Globecom Workshop (IEEE Globecom Workshop)
-
O2Fair Use and Innovation in Unlicensed Wireless Spectrum: LTE unlicensed and Wi-Fi in the 5 GHz unlicensed bandIEEE-USA Journal of Technology and Public Policy
-
T1Downlink and Uplink User Association in Dense Next-Generation Wireless NetworksBachelors Thesis, University of Texas at Austin.
-
O1Use of Electroencephalography and Galvanic Skin Response in the Prediction of an Attentive Cognitive StateHealth and Human Performance Research Summit, Dayton, CO
Working Papers
-
W8The Forest from the Trees: Lessons on Public-Sector Deployment from an Academic Collaboration with NYC Department of Parks and Recreation
-
W7What’s Wrong With Surveillance Pricing?California Law Review, 2027 (Pending official Editorial Board selection)
-
W2The Subjectivity of Monoculture
Journal Articles
-
2026 J15Connecting Application Behavior to Undermatching in New York City School ChoiceNature Cities, Accepted
-
2026 J14Combating Gerrymandering with Ranked Choice Voting: An Experimental Analysis of Multimember Districts in the United StatesOperations ResearchConference version published in ACM Conference on Economics and Computation (EC‘22) Cornell Chronicle, San Francisco Chronicle.
-
2026 J13Dropping Standardized Testing for Admissions Trades Off Information and AccessManagement ScienceConference version published in ACM Conference on Fairness, Accountability, and Transparency (FAccT‘21); also appeared in EAAMO‘21 and in the 2021 NBER Decentralization Conference ACM EAAMO Student Paper Award, 2021 INFORMS DEI Best Student Paper Award, 2022 Yale Insights.
-
2026 J12Improving flood detection with large-scale dashboard camera dataNature Communications
-
2026 J11Simpler Than You Think: The Practical Dynamics of Ranked Choice VotingJournal of Computational Social Science
-
2025 J10Inferring fine-grained migration patterns across the United StatesNature CommunicationsAssociation of American Geographers Best Student Paper Award, 2026 (Awardee: Gabriel Agostini) IPUMS Spatial Student Research Award, 2026 (Awardee: Gabriel Agostini) Cornell Chronicle.
-
2025 J9"Shopping Around": An Experiment in Preferences and Incentives for Placing Long-term PatientsProceedings of the ACM on Human-Computer InteractionJournal Track for 28th ACM SIGCHI Conference on Computer-Supported Cooperative Work & Social Computing (CSCW‘25) Impact Recognition at CSCW ‘25 Cornell Chronicle.
-
2025 I1Heterogeneous participation and allocation skews: when is choice “worth it”?ACM SIGecom ExchangesInvited; lightly reviewed by editors
-
2025 J8Faster Information for Effective Long-Term Discharge: A Field Study in Adult Foster CareProceedings of the ACM on Human-Computer InteractionJournal Track for 28th ACM SIGCHI Conference on Computer-Supported Cooperative Work & Social Computing (CSCW‘25). Also appeared in EAAMO‘23 Best Paper Award, and Diversity & Inclusion Recognition at CSCW ‘25 Cornell Chronicle.
-
2025 J7Addressing Discretization-Induced Bias in Demographic PredictionPNAS NexusConference version appeared in ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT 2024).
-
2023 J6Quantifying Spatial Under-reporting Disparities in Resident CrowdsourcingNature Computational ScienceConference version published in ACM Conference on Economics and Computation (EC‘22), titled “Equity in Resident Crowdsourcing: Measuring Under-reporting without Ground Truth Data” INFORMS Junior Faculty Interest Group Paper Award (Finalist), 2023 Cornell Chronicle.
-
2022 J5Driver Surge PricingManagement ScienceConference version published in EC‘20.
-
2021 J4Designing Informative Rating Systems: Evidence from an Online Labor MarketManufacturing & Service Operations ManagementNew York Times, Stanford Engineering magazine. M&SOM student paper award (2nd place), 2020 Conference version published in EC‘20.
-
2020 J3Markets for Public Decision-makingSocial Choice and WelfareConference version published in WINE‘18.
-
2019 J2Iterative Local Voting for Collective Decision-making in Continuous SpacesJournal of Artificial Intelligence Research (JAIR)Conference version published in WWW‘17.
-
2018 J1Word Embeddings Quantify 100 Years of Gender and Ethnic StereotypesProceedings of the National Academy of Sciences (PNAS)
Peer Reviewed Conference Proceedings (without journal versions)
-
2026 C34Measuring the Impact of Urban Cycling Infrastructure with Large Scale E-Bike GPS DataEAAMO‘26Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO‘26)
-
2026 C33Personalized Recommendations without Inducing Congestion: Mitigating Disparities in the NYC High School MatchEC‘26ACM Conference on Economics and Computation (EC‘26)EC Best Paper with Student Lead Author, 2026 (2 awards out of 1,111 submissions) INFORMS Doing Good with Good OR Student Paper Award, 2026 (Finalist; winners TBA) Plenary talk at IC2S2‘26 (top 2.3% of submissions)
-
2026 C32Position: Use Sparse Autoencoders to Discover UnknownsICML‘26 Position Paper TrackInternational Conference on Machine Learning (ICML‘26 Position Paper Track)
-
2026 C31How Many Features Can a Language Model Store Under the Linear Representation Hypothesis?COLT‘26Conference on Learning Theory (COLT‘26)Spotlight Talk at ICML Workshop on Compositional Learning (3 of 210 submissions)
-
2026 C30Urban Incident Prediction with Graph Neural Networks: Integrating Government Ratings and Crowdsourced ReportsAAAI‘26AAAI Conference on Artificial Intelligence (AAAI‘26)
-
2026 C29Capacity Constraints Make Admissions Processes Less PredictableAAAI‘26AAAI Conference on Artificial Intelligence (AAAI‘26)Oral presentation.
-
2025 C28Correlated Errors in Large Language ModelsICML‘25International Conference on Machine Learning (ICML‘25)
-
2025 C27Sparse Autoencoders for Hypothesis GenerationICML‘25International Conference on Machine Learning (ICML‘25)
-
2025 C26Learning Disease Progression Models That Capture Health DisparitiesCHIL ‘25Conference on Health, Inference, and Learning (CHIL ‘25)Best Paper Award for Impact & Society at CHIL ‘25
-
2025 C25Balancing Producer Fairness and Efficiency via Prior-Weighted Rating System DesignICWSM ‘25International AAAI Conference on Web and Social Media (ICWSM ‘25)
-
2025 C24A No Free Lunch Theorem for Human-AI CollaborationAAAI‘25AAAI Conference on Artificial Intelligence (AAAI‘25)
-
2024 C23Monoculture in Matching MarketsNeurIPS ‘24Neural Information Processing Systems (NeurIPS ‘24)
-
2024 C22User-item fairness tradeoffs in recommendationsNeurIPS ‘24Neural Information Processing Systems (NeurIPS ‘24)
-
2024 C21Ending Affirmative Action Harms Diversity Without Improving Academic MeritEAAMO ’24Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO ’24)
-
2024 C20Wisdom and Foolishness of Noisy Matching MarketsEC‘24ACM Conference on Economics and Computation (EC‘24)
-
2024 C19Equitable Congestion Pricing under the Markovian Traffic Model: An Application to BogotaEC‘24ACM Conference on Economics and Computation (EC‘24)
-
2024 C18Topics, Authors, and Institutions in Large Language Model Research: Trends from 17K arXiv PapersNAACL‘24The North American Chapter of the Association for Computational Linguistics (NAACL‘24)
-
2024 C17Domain constraints improve risk prediction when outcome data is missingICLR‘24International Conference on Learning Representations (ICLR‘24)
-
2024 C16Reconciling the accuracy-diversity trade-off in recommendationsWWW‘24The ACM Web Conference (WWW‘24)Oral Presentation
-
2024 C15A Bayesian Spatial Model to Correct Under-Reporting in Urban CrowdsourcingAAAI‘24AAAI Conference on Artificial Intelligence (AAAI‘24)Oral Presentation
-
2024 C14Identifying and Addressing Disparities in Public Libraries with Bayesian Latent Variable ModelingAAAI‘24AAAI Conference on Artificial Intelligence (AAAI‘24)
-
2023 C13Supply-Side Equilibria in Recommender SystemsNeurIPS ‘23Neural Information Processing Systems (NeurIPS ‘23)
-
2023 C12Interface Design to Mitigate Inflation in Recommender SystemsRecSys ’23 Short paperACM Conference on Recommender Systems (RecSys ’23 Short paper)
-
2023 C11Coarse race data conceals disparities in clinical risk score performanceML4HC‘23Machine Learning for Healthcare (ML4HC) (ML4HC‘23)ML4H Best Findings Paper (Honorable Mention), 2023 Cornell Chronicle
-
2022 C10Fair ranking: a critical review, challenges, and future directionsFAccT‘22ACM Conference on Fairness, Accountability, and Transparency (FAccT‘22)This work was written as part of a distributed, student-led working group of Mechanism Design for Social Good
-
2022 C9Trucks Don’t Mean Trump: Diagnosing Human Error in Image AnalysisFAccT‘22ACM Conference on Fairness, Accountability, and Transparency (FAccT‘22)
-
2022 C8Strategic RankingAISTATS‘22International Conference on Artificial Intelligence and Statistics (AISTATS‘22)
-
2021 C7Test-optional Policies: Overcoming Strategic Behavior and Informational GapsEAAMO‘21AAAI/ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO‘21)
-
2021 C6The Stereotyping Problem in Collaboratively Filtered Recommender SystemsEAAMO‘21AAAI/ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO‘21)
-
2020 C5Fair Allocation through Selective Information AcquisitionAIES‘20AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES‘20)
-
2019 C4Who is in Your Top Three? Optimizing Learning in Elections with Many CandidatesHCOMP‘19AAAI Conference on Human Computation and Crowdsourcing (HCOMP‘19)
-
2019 C3Analyzing Polarization in Social Media: Method and Application to Tweets on 21 Mass ShootingsNAACL‘19Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL‘19)
-
2019 C2Designing Optimal Binary Rating SystemsAISTATS‘19International Conference on Artificial Intelligence and Statistics (AISTATS‘19)
-
2015 C1Impact of Dual Slope Path Loss on User Association in HetNetsIEEE Globecom WorkshopIEEE Globecom Workshop (IEEE Globecom Workshop)
Other (workshops and technical reports)
-
2025 O6Optimizing Library Usage and Browser Experience: Application to the New York Public Library
-
2024 O5Choosing the Right Weights: Balancing Value, Strategy, and Noise in Recommender Systems
-
2019 O4Deliberative Democracy with the Online Deliberation PlatformHCOMP‘19 DemoAAAI Conference on Human Computation and Crowdsourcing Demo Track (HCOMP‘19 Demo)
-
2018 O3Comparing Voting Methods for Budget Decisions on the ASSU Ballot
-
2015 O2Fair Use and Innovation in Unlicensed Wireless Spectrum: LTE unlicensed and Wi-Fi in the 5 GHz unlicensed bandIEEE-USA Journal of Technology and Public Policy
-
2015 O1Use of Electroencephalography and Galvanic Skin Response in the Prediction of an Attentive Cognitive StateHealth and Human Performance Research Summit, Dayton, CO
Theses
-
2020 T2Designing Marketplaces and Civic Engagement Platforms: Learning, Incentives, and PricingPhD Dissertation, Stanford University INFORMS George Dantzig Dissertation Award, 2020 ACM SIGecom dissertation award (Honorable mention), 2021
-
2015 T1Downlink and Uplink User Association in Dense Next-Generation Wireless NetworksBachelors Thesis, University of Texas at Austin.
Contact me at ngarg@REMOVETHIScornell.REMOVETHISedu. Applicants: please read the information at the Contact page before emailing me.