Lucas Monteiro Paes

Lucas Monteiro Paes

Auto-generated profile, based on publicly available information

Lucas Monteiro Paes

School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 June 2024
Verified
Conference Paper
Open access
JGJuan Felipe Gomez
Juan Felipe Gomez
CMCaio Machado
Caio Machado
LPLucas Monteiro Paes
Lucas Monteiro Paes
FCFlavio Du Pin Calmon
Flavio Du Pin Calmon
Machine learning (ML) is widely used to moderate online content. Despite its scalability relative to human moderation, the use of ML introduces unique challenges to content moderation. One such challenge is predictive multiplicity: multiple competing models for content classification may perform equally well on average, yet assign conflicting predictions to the same content. This multiplicity can result from seemingly innocuous choices made during training, which do not meaningfully change the accuracy of the ML model, but can nevertheless change what the model gets wrong. We experimentally demonstrate how content moderation tools can arbitrarily classify samples as “toxic,” leading to arbitrary restrictions on speech. We use the principles set by the International Covenant on Civil and Political Rights (ICCPR), namely freedom of expression, non-discrimination, and procedural justice to interpret the effects of these findings in terms of Human Rights. We analyze (i) the extent of predictive multiplicity among popular state-of-the-art LLMs used for detecting “toxic” content; (ii) the disparate impact of this arbitrariness across social groups; and (iii) the magnitude of model multiplicity on content that is unanimously recognized as toxic by human annotators. Our findings indicate that the up-scaled algorithmic moderation risks legitimizing an “algorithmic leviathan”, where an algorithm disproportionately manages human rights. To mitigate such risks, our study underscores the need to identify and increase the transparency of arbitrariness in content moderation applications. Our findings have implications to content moderation and intermediary liability laws being discussed and passed in many countries, such as the Digital Services Act in the European Union, the Online Safety Act in the United Kingdom, and the recent TSE resolutions in Brazil.

23 June 2025
Verified
Conference Paper
Open access
MBMaarten Buyl
Maarten Buyl
HKHadi Khalaf
Hadi Khalaf
CVClaudio Mayrink Verdun
Claudio Mayrink Verdun
LPLucas Monteiro Paes
Lucas Monteiro Paes
CMCaio Cesar Vieira Machado
Caio Cesar Vieira Machado
FCFlavio Du Pin Calmon
Flavio Du Pin Calmon

9 May 2024 in IEEE Journal on Selected Areas in Information Theory
Verified
Journal Article
Closed access
LPLucas Monteiro Paes
Lucas Monteiro Paes
ASAnanda Theertha Suresh
Ananda Theertha Suresh
ABAlex Beutel
Alex Beutel
FCFlavio Du Pin Calmon
Flavio Du Pin Calmon
ABAhmad Beirami
Ahmad Beirami
Machine learning (ML) models used in prediction and classification tasks may display performance disparities across population groups determined by sensitive attributes (e.g., race, sex, age). We consider the problem of evaluating the performance of a fixed ML model across population groups defined by multiple sensitive attributes (e.g., race and sex and age). Here, the sample complexity for estimating the worst-case performance gap across groups (e.g., the largest difference in error rates) increases exponentially with the number of group-denoting sensitive attributes. To address this issue, we propose an approach to test for performance disparities based on Conditional Value-at-Risk (CVaR). By allowing a small probabilistic slack on the groups over which a model has approximately equal performance, we show that the sample complexity required for discovering performance violations is reduced exponentially to be at most upper bounded by the square root of the number of groups. As a byproduct of our analysis, when the groups are weighted by a specific prior distribution, we show that Rényi entropy of order 2/3 of the prior distribution captures the sample complexity of the proposed CVaR test algorithm. Finally, we also show that there exists a non-i.i.d. data collection strategy that results in a sample complexity independent of the number of groups.