Lucas H. McCabe

Present:

I am a Data Science Fellow at LMI Consulting and a computer science PhD candidate at The George Washington University, advised by H. Howie Huang. My work primarily pertains to natural language processing and computational social science; see my research page for more.

I am grateful to have been recognized as a DARPA Riser, Luminary Awardee (LMI), and Bernstein Scholar (Institute for Quantitative Biomedicine, formerly “BioMaPS”).

Past:

I completed my master’s at Johns Hopkins under Tom Woolf. Before that, I studied math and computer science at Rutgers and worked in the Moghe Lab.

selected publications

Appl. Netw. Sci.
Network analysis of U.S. non-fatal opioid-involved overdose journeys, 2018–2023

L. H. McCabe, N. Masuda, S. Casillas, N. Danneman, A. Alic, and R. Law

Applied Network Science, Nov 2024

Abs Bib HTML PDF

We present a nation-wide network analysis of non-fatal opioid-involved overdose journeys in the United States. Leveraging a unique proprietary dataset of Emergency Medical Services incidents, we construct a journey-to-overdose geospatial network capturing nearly half a million opioid-involved overdose events spanning 2018-2023. We analyze the structure and sociological profile of the nodes, which are counties or their equivalents, characterize the distribution of overdose journey lengths, and investigate changes in the journey network between 2018 and 2023. Our findings include that authority and hub nodes identified by the HITS algorithm tend to be located in urban areas and involved in overdose journeys with particularly long geographical distances.
@article{mccabe2024overdosenetworks, title = {Network analysis of U.S. non-fatal opioid-involved overdose journeys, 2018–2023}, author = {McCabe, L. H. and Masuda, N. and Casillas, S. and Danneman, N. and Alic, A. and Law, R.}, journal = {Applied Network Science}, month = nov, year = {2024}, doi = {https://doi.org/10.1007/s41109-024-00661-z}, }
EMNLP
Prompts have evil twins

R. Melamed, L. H. McCabe, T. Wakhare, Y. Kim, H. H. Huang, and E. Boix-Adsera

In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing , Nov 2024

Abs Bib HTML PDF Code

We discover that many natural-language prompts can be replaced by corresponding prompts that are unintelligible to humans but that provably elicit similar behavior in language models. We call these prompts “evil twins” because they are obfuscated and uninterpretable (evil), but at the same time mimic the functionality of the original natural-language prompts (twins). Remarkably, evil twins transfer between models. We find these prompts by solving a maximum-likelihood problem which has applications of independent interest.
@inproceedings{melamed2023propane, title = {Prompts have evil twins}, author = {Melamed, R. and McCabe, L. H. and Wakhare, T. and Kim, Y. and Huang, H. H. and Boix-Adsera, E.}, editor = {Al-Onaizan, Yaser and Bansal, Mohit and Chen, Yun-Nung}, booktitle = {Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing}, month = nov, year = {2024}, address = {Miami, Florida, USA}, publisher = {Association for Computational Linguistics}, pages = {46--74}, }