SFSorelle A. Friedler

Sorelle A. Friedler

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Sorelle A. Friedler has a total of 90 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Popformer: Learning general signatures of positive selection with a self-supervised transformer

    by Leon Zong, Sorelle A. Friedler, Sara Mathieson

    Understanding natural selection can help shed light on the genetics underpinning adaptive evolution. The widespread availability of large-scale human genetic variation data has led to the development of data-driven methods for detecting signatures of selection, many of which are based on deep learning. However, these methods often fail to generalize well to the diversity of selection signatures across a broad range of evolutionary scenarios. We propose a novel transformer-based model, Popformer, for learning encodings of general patterns of genetic variation. Popformer includes site-wise and haplotype-wise attention, allowing us to capture variation among both genetic positions and individuals. It additionally learns relative positional embeddings for inter-SNP distances. The model is pre-trained with an analog of the masked language modeling objective across a range of real human genomic data, similar to the task of genetic imputation. Using dimensionality reduction, we show that the pre-trained model learns meaningful embeddings of genomic windows that correspond with population structure. We also show that the model can accurately perform genotype imputation. We further fine-tune the model on selection classification and demonstrate that our model is more accurate than other selection classification methods, on selection simulations of both well-specified and mis-specified demographic models. Using a novel real data validation approach, we apply Popformer to human data from the 1000 Genomes Project and reveal its ability to generalize from simulated data to the diversity of real data. Overall, our method provides a new direction for population genetic inference methods, with future fine-tuning applications including the inference of recombination rates, introgression, and local ancestry.

  • Open Access

    Do Language Models Pass the Bechdel Test? Auditing Gender Biases in LLM-Generated Screenplays

    by Megha N. Govindu, Stephanie T. Wang, Sorelle A. Friedler, Danaé Metaxa

  • Open Access

    Edge interventions can mitigate demographic and prestige disparities in the computer science coauthorship network

    by Kate Barnes, Mia Ellis-Einhorn, Carolina Chavez, Nayera Hasan, Mohammad Fanous, Blair D. Sullivan, Sorelle Friedler, Aaron Clauset

  • Open Access

    Self-Promotion in LLM Recommendations

    by Rustom M. Dubash, Sorelle A. Friedler


  • Open Access

    Identity-related Speech Suppression in Generative AI Content Moderation

    by Grace Proebsting, Oghenefejiro Isaacs Anigboro, Charlie M. Crawford, Danaé Metaxa, Sorelle A. Friedler


  • Open Access

    Edge interventions can mitigate demographic and prestige disparities in the Computer Science coauthorship network

    by Kate Barnes, Mia Ellis-Einhorn, Carolina Chávez-Ruelas, Nayera Hasan, Mohammad Fanous, Blair D. Sullivan, Sorelle Friedler, Aaron Clauset

    Social factors such as demographic traits and institutional prestige structure the creation and dissemination of ideas in academic publishing. One place these effects can be observed is in how central or peripheral a researcher is in the coauthorship network. Here we investigate inequities in network centrality in a hand-collected data set of 5,670 U.S.-based faculty employed in Ph.D.-granting Computer Science departments and their DBLP coauthorship connections. We introduce algorithms for combining name- and perception-based demographic labels by maximizing alignment with self-reported demographics from a survey of faculty from our census. We find that women and individuals with minoritized race identities are less central in the computer science coauthorship network, implying worse access to and ability to spread information. Centrality is also highly correlated with prestige, such that faculty in top-ranked departments are at the core and those in low-ranked departments are in the peripheries of the computer science coauthorship network. We show that these disparities can be mitigated using simulated edge interventions, interpreted as facilitated collaborations. Our intervention increases the centrality of target individuals, chosen independently of the network structure, by linking them with researchers from highly ranked institutions. When applied to scholars during their Ph.D., the intervention also improves the predicted rank of their placement institution in the academic job market. This work was guided by an ameliorative approach: uncovering social inequities in order to address them. By targeting scholars for intervention based on institutional prestige, we are able to improve their centrality in the coauthorship network that plays a key role in job placement and longer-term academic success.


  • Open Access

    Auditing GPT's Content Moderation Guardrails: Can ChatGPT Write Your Favorite TV Show?

    by Yaaseen Mahomed, Charlie M. Crawford, Sanjana Gautam, Sorelle A. Friedler, Danaë Metaxa

    Large language models (LLMs) are increasingly appearing in consumer-facing products. To prevent problematic use, the organizations behind these systems have put content moderation guardrails in place that prevent the models from generating content they consider harmful. However, most of these enforcement standards and processes are opaque. Although they play a major role in the user experience of these tools, automated content moderation tools have received relatively less attention than other aspects of the models. This study undertakes an algorithm audit of OpenAI’s ChatGPT with the goal of better understanding its content moderation guardrails and their potential biases. To evaluate performance on a broad cultural range of content, we generate a dataset of 100 popular United States television shows with one to three synopses for each episode in the first season of each show (3,309 total synopses). We probe GPT’s content moderation endpoint (ME) to identify violating content both in the synopses themselves, and in GPT’s own outputs when asked to generate a script based on each synopsis, also comparing with ME outputs on 81 real scripts from the same TV shows (269,578 total ME outputs). Our findings show that a large number of GPT-generated and real scripts flag as content violations (about 18% of GPT scripts and 69% of real ones). Using metadata, we find that TV maturity ratings, as well as certain genres (Animation, Crime, Fantasy, and others) are statistically significantly related to a script’s likelihood of flagging. We conclude by discussing the implications of LLM self-censorship and directions for future research on their moderation procedures.


  • Open Access

    Information access representations and social capital in networks

    by Ashkan Bashardoust, Hannah C. Beilinson, Sorelle A. Friedler, Jiajie Ma, Jade Rousseau, Carlos E. Scheidegger, Blair D. Sullivan, Nasanbayar Ulzii-Orshikh, Suresh Venkatasubramanian

    Social network position confers power and social capital. In the setting of online social networks that have massive reach, creating mathematical representations of social capital is an important step towards understanding how network position can differentially confer advantage to different groups and how network position can itself be a source of advantage. In this paper, we use well established models for information flow on networks as a base to propose a formal descriptor of the network position of a node as represented by its information access. Combining these descriptors allows a full representation of social capital across the network. Using real-world networks, we demonstrate that this representation allows the identification of differences between groups based on network specific measures of inequality of access.

  • Open Access

    Reducing Access Disparities in Networks using Edge Augmentation✱

    by Ashkan Bashardoust, Sorelle Friedler, Carlos Scheidegger, Blair D. Sullivan, Suresh Venkatasubramanian

    In social networks, a node’s position is, in and of itself, a form of social capital. Better-positioned members not only benefit from (faster) access to diverse information, but innately have more potential influence on information spread. Structural biases often arise from network formation, and can lead to significant disparities in information access based on position. Further, processes such as link recommendation can exacerbate this inequality by relying on network structure to augment connectivity. In this paper, we argue that one can understand and quantify this social capital through the lens of information flow in the network. In contrast to prior work, we consider the setting where all nodes may be sources of distinct information, and a node’s (dis)advantage takes into account its ability to access all information available on the network, not just that from a single source. We introduce three new measures of advantage (broadcast, influence, and control), which are quantified in terms of position in the network using access signatures – vectors that represent a node’s ability to share information with each other node in the network. We then consider the problem of improving equity by making interventions to increase the access of the least-advantaged nodes. Since all nodes are already sources of information in our model, we argue that edge augmentation is most appropriate for mitigating bias in the network structure, and frame a budgeted intervention problem for maximizing broadcast (minimum pairwise access) over the network. Finally, we propose heuristic strategies for selecting edge augmentations and empirically evaluate their performance on a corpus of real-world social networks. We demonstrate that a small number of interventions can not only significantly increase the broadcast measure of access for the least-advantaged nodes (over 5 times more than random), but also simultaneously improve the minimum influence. Additional analysis shows that edge augmentations targeted at improving minimum pairwise access can also dramatically shrink the gap in advantage between nodes (over ) and reduce disparities between their access signatures.


  • Open Access

    Measuring and mitigating voting access disparities: a study of race and polling locations in Florida and North Carolina

    by Mohsen Abbasi, Calvin Barrett, Kristian Lum, Sorelle A. Friedler, Suresh Venkatasubramanian

    Voter suppression and associated racial disparities in access to voting are long-standing civil rights concerns in the United States. A history of violent explicit discouragement has shifted to more subtle access limitations that can include long lines and wait times, long travel times to reach a polling station, and other logistical barriers to voting. Our focus in this work is on quantifying disparities in voting access pertaining to the overall time-to-vote, and how they could be remedied via a better choice of polling location or provisioning more sites where voters can cast ballots. However, appropriately calibrating access disparities is difficult because of the need to account for factors such as population density and different community expectations for reasonable travel times. In this paper, we perform one of the first large-scale studies of voter access to polling locations, using real-world voter data from Florida and North Carolina in the 2020 general election. We develop a methodology for the calibrated measurement of disparities in polling location "load" and distance to polling locations based on a novel normalized distance metric to model the voter experience of distance. We find that voter turnout is reduced when this normalized distance to polling locations increases, and that non-white voters had to travel further to the polls in Florida (using this normalized distance) than White voters. We also introduce algorithms, with modifications to handle scale, that can reduce these disparities by suggesting new polling locations from a given list of identified public locations (including schools and libraries). The developed voting access measurement methodology and algorithmic remediation technique demonstrates that better polling location placement is possible.