I am an Assistant Professor in the University of South Florida's Bellini College of Artificial Intelligence, Cybersecurity and Computing and Director of the Language GRASP Lab. I received my doctorate at the University of Rochester, advised by Len Schubert. My research interests are in the intersection of natural language processing, computer reasoning, and computational linguistics. I create algorithms that model formal representations of natural language meaning and structured information to enable computers to interact with language in a generalizable manner. My lab now also uses formal meaning representations for neurosymbolic data augmentation, generating training data that is structurally valid by construction for tasks such as semantic parsing, where human annotation is scarce. More recently, we have also been working on trustworthy AI, studying the safety and reliability of large language models across diverse social and cultural settings, as well as on event detection, and I collaborate on building systems that generate educational games from sources that teachers select. Now that large language models use natural language fluently across a wide range of settings, my long-term research goal is to make the reasoning behind that language use reliable and open to verification, so that people can tell when an AI system's conclusions deserve their trust.

Email: genekim AT usf DOT edu
Office: Engineering Building II 376

CV
PhD Thesis: Corpus Annotation, Parsing, and Inference for Episodic Logic Type Structure

Students

PhD Students

• Mahammed KamruzzamanSafety and trustworthiness of generative AI
• Abdullah Al MonsurAI in education, building on event detection
• Jeremy RutterNatural language constraints for reinforcement learningCo-advised with Yu Sun
• Shrabon Kumar DasNeurosymbolic data augmentation

Master's Students

• Hieu Minh NguyenSycophancy in LLMs

Undergraduate Researchers

• Aviral GuptaHonors thesis
• Bao Gia KhuongUR2PhD, Fall 2026
• Riyon PraveenUR2PhD, Fall 2026

Alumni

• Kenneth VerzylMS student, 2025–2026; undergraduate researcher, 2024–2025Thesis: Finding Reductions in Connected Speech Using Masked Language Models [ProQuest]
• Dayne GuyMS student, 2025–2026; undergraduate researcher, 2025
• Nitesh Vamshi BommisettyMS student, 2024–2025Now at JPMorgan ChaseThesis: Zero-Shot and Few-Shot Event Extraction With Large Language Models Across Long-Tailed Distribution [ProQuest]
• Sai Durga Shashidhar PothukuchiMS student, 2024Built the LGRASP Lab Website
• Maneesh Reddy ChamakuraMS student, 2022–2024Now an SDE at Amazon Web Services
• Anubhav GuptaUndergraduate researcher (UR2PhD), Spring 2026
• Otavio Moraes RosaUndergraduate researcher (UR2PhD), Spring 2026
• Farhan Shahriar AbrarUndergraduate researcher (UR2PhD), Fall 2025
• Ishraq KhanUndergraduate researcher (UR2PhD), Fall 2025
• Hieu Trung NguyenUndergraduate researcher (UR2PhD), Fall 2025
• Jason RudinskyUndergraduate researcher, 2024–2025
• Fiorella Andrea Ratti TamayoUndergraduate researcher, 2024
• Adam BrohlUndergraduate researcher, 2023–2024 (Knox College)Now a PhD student at the University of Rochester
• Justin DelgadoUndergraduate researcher, 2023–2024
• Suryakanth Prasad GottipatiUndergraduate researcher, 2022–2024
• Omar AbdulrahmanUndergraduate researcher, 2022–2023
• Taylor TiedgeUndergraduate researcher, 2022

Online Resources

Trustworthy AI

• Subtler Biases Benchmark
  A templated benchmark of ageism, beauty, institutional, and nationality biases in generative models, with code to extend it, from Findings of ACL 2024.
• Cognitively Motivated Prompting Techniques for Reducing Social Bias
  Data and code for Prompting Techniques for Reducing Social Bias in LLMs through System 1 and System 2 Cognitive Processes, from RANLP 2025.
• Brand Bias in LLMs
  Data and evaluation code for “Global is Good, Local is Bad?”: Understanding Brand Bias in LLMs, from EMNLP 2024.
• Nationality-Assigned Personas
  Data and code for Exploring Changes in Nation Perception with Nationality-Assigned Personas in LLMs, from EMNLP 2025.
• BanStereoSet
  A dataset of 1,194 Bangla examples for measuring stereotypical social biases in LLMs, from Findings of ACL 2025.
• Name Age Perception
  Data for The Impact of Name Age Perception on Job Recommendations in LLMs, from Findings of ACL 2025.
• BanMANI
  A dataset for identifying manipulated social media news in Bangla, from the 2023 BUCC workshop.
• Disability Disclosure in Candidate Selection
  Data for The Impact of Disability Disclosure on Fairness and Bias in LLM-Driven Candidate Selection, from FLAIRS 2025.

Other Recent Work

• Context-Aware Event Detection
  Code for Event Detection with a Context-Aware Encoder and LoRA for Improved Performance on Long-Tailed Classes, from Findings of EACL 2026.
• Resource-Aware Sentiment Analysis
  Data and code for Efficient Sentiment Analysis: A Resource-Aware Evaluation of Feature Extraction Techniques, Ensembling, and Deep Learning Models, from SocialNLP 2023.

Unscoped Episodic Logical Form (ULF)

• Project Website
  This is the general project website. It holds a collection of blog posts that describe ULF in more detail than possible in publications or focused documents. It also links to resources used by the project, such as annotation guidelines and code.
• Annotation Guidelines (Version 1.0.1)
  A tutorial on annotating sentences with ULFs, which annotators used for training and as a reference throughout annotation. This is the version that accompanies my dissertation.
• Annotated Dataset, Release 1.0
  The release contains 1,738 hand-annotated sentences (1,378 train, 180 dev, and 180 test), drawn from crowd-sourced translations, news text, a question dataset, and novels. It accompanies the IWCS 2021 parser paper.
• Parser
  The transition-based parser from IWCS 2021.
• FraCaS Inference and Discourse Inference
  Natural logic inference on the FraCaS dataset (NALOMA 2021), and a pared-down version of the discourse inference code from DMR 2019.
• Core Library, English Generation, and Annotation Sanity Checker
  A Common Lisp library for manipulating ULFs, a system that maps ULFs to English sentences, and a quick type checker that helps annotators catch errors.

Earlier Work

• Schema-Based Story Understanding
  Lane Lawley's code for Towards Natural Language Story Understanding with Rich Logical Schemas, from NLCS 2019.
• ImplIE
  Code for the implicit relation extractor described in our 2015 technical report.
• Format String Checker
  The format string type system from ISSTA 2014, distributed with the Checker Framework.

Publications

• Demographic Prompting at Scale: When More Attributes Hurt LLM–Human Agreement
  Mahammed Kamruzzaman, Shrabon Kumar Das, and Gene Louis Kim
  In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2026 (to appear)
  
• Breaking the Benchmark: Revealing LLM Bias via Minimal Contextual Augmentation
  Kaveh Eskandari Miandoab, Mahammed Kamruzzaman, Arshia Gharooni, Gene Louis Kim, Vasanth Sarathy, and Ninareh Mehrabi
  In Proceedings of the Fifteenth Language Resources and Evaluation Conference (LREC), 2026
  
• Event Detection with a Context-Aware Encoder and LoRA for Improved Performance on Long-Tailed Classes
  Abdullah Al Monsur, Nitesh Vamshi Bommisetty, and Gene Louis Kim
  In Findings of the Association for Computational Linguistics: EACL 2026
  [Code]
• ‘A Woman is More Culturally Knowledgeable than A Man?’: The Effect of Personas on Cultural Norm Interpretation in LLMs
  Mahammed Kamruzzaman, Hieu Minh Nguyen, Nazmul Hassan, and Gene Louis Kim
  In Proceedings of the First Workshop on Multilingual Multicultural Evaluation (MME), 2026
  
• From Anger to Joy: How Nationality Personas Shape Emotion Attribution in Large Language Models
  Mahammed Kamruzzaman, Abdullah Al Monsur, Gene Louis Kim, and Anshuman Chhabra
  In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (IJCNLP-AACL), 2025
  
• Exploring Changes in Nation Perception with Nationality-Assigned Personas in LLMs
  Mahammed Kamruzzaman and Gene Louis Kim
  In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025
  [Code]
• Evaluating the Impact of Racial Cues on MLLMs Judgements of Politeness and Offensiveness
  Mahammed Kamruzzaman and Gene Louis Kim
  In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2025
  
• Prompting Techniques for Reducing Social Bias in LLMs through System 1 and System 2 Cognitive Processes
  Mahammed Kamruzzaman and Gene Louis Kim
  In Proceedings of the 15th International Conference on Recent Advances in Natural Language Processing (RANLP), 2025
  [Code]
• The Impact of Name Age Perception on Job Recommendations in LLMs
  Mahammed Kamruzzaman and Gene Louis Kim
  In Findings of the Association for Computational Linguistics: ACL 2025
  [Data]
• BanStereoSet: A Dataset to Measure Stereotypical Social Biases in LLMs for Bangla
  Mahammed Kamruzzaman, Abdullah Al Monsur, Shrabon Kumar Das, Enamul Hassan, and Gene Louis Kim
  In Findings of the Association for Computational Linguistics: ACL 2025
  [Data]
• The Impact of Disability Disclosure on Fairness and Bias in LLM-Driven Candidate Selection
  Mahammed Kamruzzaman and Gene Louis Kim
  In Proceedings of the 38th International Conference of the Florida Artificial Intelligence Research Society (FLAIRS), 2025
  [Data]
• “Global is Good, Local is Bad?”: Understanding Brand Bias in LLMs
  Mahammed Kamruzzaman, Hieu Minh Nguyen, and Gene Louis Kim
  In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024
  [Code]
• Understanding Novel, Real World Scenes: Gist Strength and the Riches Beyond
  Adison Roman, Reilly Orman, Lewis Evans, Chad Hense, Aaron Sharrett, Skyler Torres, Gene Kim, and Thomas Sanocki
  Journal of Vision, 24(10), 2024
  
• Investigating Subtler Biases in LLMs: Ageism, Beauty, Institutional, and Nationality Bias in Generative Models
  Mahammed Kamruzzaman, Md. Shovon, and Gene Kim
  In Findings of the Association for Computational Linguistics: ACL 2024
  [Data]
• Finetuning LLMs for Automatic Concept to TTI Prompt Generation (Student Abstract)
  Jeremy Rutter, Maneesh Reddy Chamakura, Justin Delgado, and Gene Louis Kim
  In Proceedings of the 38th AAAI Conference on Artificial Intelligence (AAAI), 2024
  
• Monotonic Inference with Unscoped Episodic Logical Forms: From Principles to System
  Gene Louis Kim, Mandar Juvekar, Junis Ekmekciu, Viet Duong, and Lenhart Schubert
  Journal of Logic, Language and Information, 2023
  
• Efficient Sentiment Analysis: A Resource-Aware Evaluation of Feature Extraction Techniques, Ensembling, and Deep Learning Models
  Mahammed Kamruzzaman and Gene Kim
  In Proceedings of the 11th International Workshop on Natural Language Processing for Social Media (SocialNLP), 2023
  [Code]
• BanMANI: A Dataset to Identify Manipulated Social Media News in Bangla
  Mahammed Kamruzzaman, Md. Minul Islam Shovon, and Gene Kim
  In Proceedings of the Workshop on Computational Terminology in NLP and Translation Studies (ConTeNTS) Incorporating the 16th Workshop on Building and Using Comparable Corpora (BUCC), 2023
  [Data]
• Semantically Informed Data Augmentation for Unscoped Episodic Logical Forms
  Mandar Juvekar, Gene Louis Kim, and Lenhart Schubert
  In Proceedings of the 15th International Conference on Computational Semantics (IWCS), 2023
  
• A Transition-based Parser for Unscoped Episodic Logical Forms
  Gene Louis Kim, Viet Duong, Xin Lu, and Lenhart Schubert
  In Proceedings of the 14th International Conference on Computational Semantics (IWCS), 2021
  [Code] [Data]
• A (Mostly) Symbolic System for Monotonic Inference with Unscoped Episodic Logical Forms
  Gene Louis Kim, Mandar Juvekar, Junis Ekmekciu, Viet Duong, and Lenhart Schubert
  In Proceedings of the 2nd Workshop on Natural Logic Meets Machine Learning (NALOMA), 2021
  [Code]
• Montague Grammar Induction
  Gene Louis Kim and Aaron Steven White
  In Proceedings of Semantics and Linguistic Theory 30 (SALT 30), 2021
  [DOI]
• Monotonic Inference for Underspecified Episodic Logic
  Gene Louis Kim, Mandar Juvekar, and Lenhart Schubert
  In Proceedings of the 1st Workshop on Natural Logic Meets Machine Learning (NALOMA), 2020
  [Slides] [Bibtex]
• Generating Discourse Inferences from Unscoped Episodic Logical Formulas
  Gene Louis Kim, Benjamin Kane, Viet Duong, Muskaan Mendiratta, Graeme McGuire, Sophie Sackstein, Georgiy Platonov, and Lenhart Schubert
  In Proceedings of the 1st Workshop on Designing Meaning Representations (DMR), 2019
  [Code] [Slides] [Bibtex]
• Towards Natural Language Story Understanding with Rich Logical Schemas
  Lane Lawley, Gene Louis Kim, and Lenhart Schubert
  In Proceedings of the Sixth Workshop on Natural Language and Computer Science (NLCS), 2019
  [Code] [Slides] [Bibtex]
• A Type-coherent, Expressive Representation as an Initial Step to Language Understanding
  Gene Louis Kim and Lenhart Schubert
  In Proceedings of the 13th International Conference on Computational Semantics (IWCS), 2019
  [Annotation Guidelines] [Slides] [Bibtex]
• Towards Parsing Unscoped Episodic Logical Forms with a Cache Transition Parser (Poster Abstract)
  Gene Louis Kim
  In Proceedings of the 32nd International Conference of the Florida Artificial Intelligence Research Society (FLAIRS), 2019
  [Poster]
• Intension, Attitude, and Tense Annotation in a High-Fidelity Semantic Representation
  Gene Kim and Lenhart Schubert
  In Proceedings of the Workshop on Computational Semantics Beyond Events and Roles (SemBEaR), 2017
  [Slides] [Bibtex]
• High-Fidelity Lexical Axiom Construction from Verb Glosses
  Gene Kim and Lenhart Schubert
  In Proceedings of the Fifth Joint Conference on Lexical and Computational Semantics (*SEM), 2016
  [Supplementary Document] [Slides] [Bibtex]
• A Type System for Format Strings
  Konstantin Weitz, Gene Kim, Siwakorn Srisakaokul, and Michael D. Ernst
  In Proceedings of the 2014 International Symposium on Software Testing and Analysis (ISSTA), 2014
  [Code] [Bibtex]
• A Format String Checker for Java
  Konstantin Weitz, Siwakorn Srisakaokul, Gene Kim, and Michael D. Ernst
  In Proceedings of the 2014 International Symposium on Software Testing and Analysis (ISSTA), 2014
  [Code] [Bibtex]

Technical Reports/Unrefereed Articles

• Token-Level Diagnosis of Sycophancy in LLMs with Attribution-Guided Steering
  Hieu Minh Nguyen, Mahammed Kamruzzaman, Anshuman Chhabra, and Gene Louis Kim
  arXiv preprint arXiv:2607.28906, 2026
  
• Thesis Proposal
  Gene Louis Kim
  Committee: Lenhart Schubert (supervisor), James Allen, Dan Gildea, Scott Grimm, Aaron Steven White
  [Slides]
• University of Washington System for 2015 KBP Cold Start Slot Filling
  Stephen Soderland, Natalie Hawkins, Gene Kim, and Daniel S. Weld
  In Proceedings of the Text Analysis Conference (TAC), 2015
  
• A Language Model for Extracting Implicit Relations
  Stephen Soderland, Gene Kim, and Natalie Hawkins
  Technical Report, 2015
  [Code]