Computer Sciences Commons^™

Structured Interactions With Llms To Support Ltl: Supplemental Information, Karenna Kung, Maanas Punuru, Alicia M. Grubb, Paola Spoletini

Computer Science: Faculty Publications

This repository contains the supplemental information for the paper: "Structured Interactions with LLMs to Support LTL", which investigates how students interact with and can be supported by LLM explanations while translating linear temporal logic formulae into English natural language.

Oompa 2025.08: A First Cut Of The Toolkit For Object-Oriented Modeling For Planning And Acting, Mark Roberts, David H. Chan, Dana S. Nau, Jamie C. Macbeth

Computer Science: Faculty Publications

OOMPA is a partially implemented Python 3.13+ toolkit for modeling hierarchical planning domains as annotated Python classes, without writing a separate PDDL or HDDL domain file. State properties, actions, and hierarchical methods attach to domain classes via decorator syntax; OOMPA projects the resulting model into a flat dictionary of dictionaries, like the Pyhop family of planners. We describe OOMPA’s motivation and architecture as we demonstrate its use in a restaurant planning domain. There are many unrealized features, so we end with a discussion of limitations and future work.

Script-Based Inferences In An Image Schema Story Understander, Jamie C. Macbeth, Boming Zhang, Sharmin Badhan

Computer Science: Faculty Publications

Recent studies of large language models (LLMs) have revealed that they lack human-like cognitive models of reasoning and understanding. An important thread of research merges image schemas into symbolic artificial intelligence systems where their use as conceptual building blocks and primitives shows promise for the study of human cognition through intelligent systems that perform neurosymbolically. The work presented in this paper demonstrates image schema primitives being used in structures of a representation system called conceptual dependency (CD) and in broader commonsense knowledge structures called scripts. We present a story under- standing system that uses image schemas as primitives in scripts …

Toward A Simplified Framework For Sequential Character Recognition, Nicholas Howe

Computer Science: Faculty Publications

This paper proposes a novel approach to handwritten charac- ter recognition using convolutional non-recurrent deep neural networks. Such a network can run in parallel at every point of a document, offer- ing potential advantages in speed over recurrent approaches. The net- work’s output feeds into a beam search optimization for final decoding. Preliminary quantitative results show that the framework can achieve bootstrap training from labeled word images. It provides an alternative to sequential models that rely on connectionist temporal classification for alignment.

Exploring The Robustness Of The Effect Of Evo On Intention Valuation Through Replication: Supplemental Material, Yesugen Baatartogtokh, Kaitlyn Cook, Alicia M. Grubb

Computer Science: Faculty Publications

Supplemental material for the paper: "Exploring the Robustness of the Effect of EVO on Intention Valuation through Replication"

Towards Connecting Requirements With Developer Artifacts In A Local Context: Supplemental Material, Sonora Halili, Karenna Kung, Paola Spoletini, Alicia M. Grubb

Computer Science: Faculty Publications

Supplemental material for the paper: "Towards Connecting Requirements with Developer Artifacts in a Local Context"

A Study Of Conceptual Primitive Elimination: Embedding Ingest Into Ptrans, Jamie C. Macbeth, Alexis Kilayko

Computer Science: Faculty Publications

In cognitive systems and cognitive linguistics, primitive decomposition systems attempt to explain cognitive phenomena by breaking things down into conceptual building blocks and provide rich and flexible representations for systems. A prime example is the Schank–Minsky Conceptual Dependency Trans-frames system, which maintains a commitment to keeping the number of primitives small and allowing them to be combined in complex ways in representing meaning, knowledge, and dynamic episodic memory. Motivated by the desire to keep the set of primitives small, this paper describes an effort to eliminate the Conceptual Dependency INGEST primitive and reconstitute its uses through combinations of the CD …

Standardizing Canine Breed Data In Veterinary Records Is Challenging, But Computer Vision Offers An Alternative Perspective On Breed Assignment, Glenvelis Perez, Yixuan He, Zihan Lyu, Yilin Chen, Nicholas Howe, Halie M. Rando

Computer Science: Faculty Publications

Dog breed is fundamental health information, especially in the context of breed-linked diseases. The standard-ization of breed terminology across health records is necessary to leverage the big data revolution for veterinary research. Breed can also inform clinical decision making. However, client-reported breeds vary in their reliability depending on how breed was determined. Surprisingly, research in computer science reports that AI can assign breed to dogs with over 90% accuracy from a photograph. Here, we explore the extent to which current research in AI is relevant to breed assignment or validation in veterinary contexts. This review provides a primer on approaches …

Augmenting, Not Replacing: The Role Of Llms In Human-Centric Formal Re: Supplemental Material, Sonora Halili, Paola Spoletini, Alicia M. Grubb

Computer Science: Faculty Publications

This repository contains the supplemental information for the RE'25 paper entitled "Augmenting, Not Replacing: The Role of LLMs in Human-Centric Formal RE" and the Smith College Departmental Honors Thesis entitled "The LTL Whisperer: Prompting AI to Explain Temporal Logic: Supplemental Information". This work investigates how and to what extent generative AI with large language models (LLMs) can assist practitioners and novices in interpreting formal requirements expressed in Linear Temporal Logic (LTL).

A View Of The Restaurant Script Through The Lens Of Hierarchical Planning, Jamie C. Macbeth, Mark Roberts, Boming Zhang, Sharmin Badhan, Molly Neu, Tanush Garg, Yining Hua, Manushaqe Muco, Mackie Zhou

Computer Science: Faculty Publications

The flexible nature of human cognition and of the structures it uses is well known, as is the difficulty of building cognitive systems that exhibit transfer and use the same structures for radically different tasks. In this paper, we perform a close examination of Schank-Abelsonian scripts, picking apart the goal- and plan- oriented nature of low-level acts and high-level reasoning inherent in them. We then view scripts through the lens of hierarchical planning systems and construct the well-known restaurant script as a hierarchical goal network planning domain. These are evidence in support of a claim that some, if not all, …

A Splash Of Color: A Dual Dive Into The Effects Of Evo On Decision-Making With Goal Models: Supplemental Material, Alicia M. Grubb, Yesugen Baatartogtokh, Irene Foster

Computer Science: Faculty Publications

Supplemental material for the paper: "A Splash of Color: A Dual Dive into the Effects of EVO on Decision-Making with Goal Models" doi.org/10.1007/s00766-024-00422-5

Missing History Of A Modern Domesticate: Historical Demographics And Genetic Diversity In Farm-Bred Red Fox Populations, Halie M. Rando, Emmarie P. Alexander, Sophie Preckler-Quisquater, Cate B. Quinn, Jeremy T. Stutchman, Jennifer L. Johnson, Estelle R. Bastounes, Beata Horecka, Kristina L. Black, Michael P. Robson, Darya V. Shepeleva, Yury E. Herbeck, Anastasiya V. Kharlamova, Lyudmila N. Trut, Jonathan N. Pauli, Benjamin N. Sacks, Anna V. Kukekova

Computer Science: Faculty Publications

The first record of captive-bred red foxes (Vulpes vulpes) dates to 1896 when a breeding enterprise emerged in the provinces of Atlantic Canada. Because its domestication happened during recent history, the red fox offers a unique opportunity to examine the genetic diversity of an emerging domesticated species in the context of documented historical and economic influences. In particular, the historical record suggests that North American and Eurasian farm-bred populations likely experienced different demographic trajectories. Here, we focus on the likely impacts of founder effects and genetic drift given historical trends in fox farming on North American and Eurasian farms. A …

Many Direct-To-Consumer Canine Genetic Tests Can Identify The Breed Of Purebred Dogs, Halie M. Rando, Kiley Graim, Greg Hampikian, Casey S. Greene

Computer Science: Faculty Publications

OBJECTIVE To compare pedigree documentation and genetic test results to evaluate whether user-provided photographs influence the breed ancestry predictions of direct-to-consumer (DTC) genetic tests for dogs.

ANIMALS 12 registered purebred pet dogs representing 12 different breeds. METHODS Each dog owner submitted 6 buccal swabs, 1 to each of 6 DTC genetic testing companies. Experimenters registered each sample per manufacturer instructions. For half of the dogs, the registration included a photograph of the DNA donor. For the other half of the dogs, photographs were swapped between dogs. DNA analysis and breed ancestry prediction were conducted by each company. The effect of …

Generative Artificial Intelligence Consensus In A Trustless Network, Edward Kim, Samu Isozaki, Naomi Sirkin, Michael Robson

Computer Science: Faculty Publications

We performed a billion locality sensitive hash comparisons between artificially generated data samples to answer the critical question - can we reproduce the results of generative AI models? Reproducibility is one of the pillars of scientific research for verifiability, bench- marking, trust, and transparency. Futhermore, we take this research to the next level by verifying the “correctness” of generative AI output in a non-deterministic, trustless, decentralized network. We generate millions of data samples from a variety of open source diffusion and large language models and describe the procedures and trade-offs between generating more verses less deterministic output. Additionally, we analyze …

Extreme Ungrading: Rewilding The Classroom Through Human-Centered Design, Johanna Brewer

Computer Science: Faculty Publications

Assessment in computer science education has grown reliant on rigid rubrics and intensive exams, a practice that yields capable yet compliant coders. In this article, I explore how we might use human-centered design to reexamine contemporary pedagogy and redesign our classrooms to cultivate a different type of programmer, one with a more critically engaged eye. Inspired by the ethos of agile development, I offer an alternative evaluation paradigm: Extreme Ungrading. Exploring results of a two-year case study applying this method to a software engineering class, this article distills actionable guidelines for enhancing learning outcomes through inclusive course development, and seeks …

Gerontovis: Data Visualization At The Confluence Of Aging, Zack While, R. Jordan Crouser, Ali Sarvghad

Computer Science: Faculty Publications

Despite the explosive growth of the aging population worldwide, older adults have been largely overlooked by visualization research. This paper is a critical reflection on the underrepresentation of older adults in visualization research. We discuss why investigating visualization at the intersection of aging matters, why older adults may have been omitted from sample populations in visualization research, how aging may affect visualization use, and how this differs from traditional accessibility research. To encourage further discussion and novel scholarship in this area, we introduce GerontoVis, a term which encapsulates research and practice of data visualization design that primarily focuses on older …

The Analyst’S Hierarchy Of Needs: Grounded Design Principles For Tailored Intelligence Analysis Tools, Antonio E. Girona, James C. Peter, Wenyuan Wang, R. Jordan Crouser

Computer Science: Faculty Publications

Intelligence analysis involves gathering, analyzing, and interpreting vast amounts of information from diverse sources to generate accurate and timely insights. Tailored tools hold great promise in providing individualized support, enhancing efficiency, and facilitating the identification of crucial intelligence gaps and trends where traditional tools fail. The effectiveness of tailored tools depends on an analyst’s unique needs and motivations, as well as the broader context in which they operate. This paper describes a series of focus discovery exercises that revealed a distinct hierarchy of needs for intelligence analysts. This reflection on the balance between competing needs is of particular value in …

Humans Against Large Language Models On Hard Paraphrase Detection Tasks, Jamie C. Macbeth, Ella Chang, Jingyu Gin Chen, Yining Hua, Sandra Grandic, Winnie X. Zheng

Computer Science: Faculty Publications

The ability to recognize that pairs or sets of language expressions “mean the same thing” is a cognitive task for which meaning representation is clearly a central issue. This paper uses the task of paraphrasing to study meaning representation in a cognitive system. The main claim is that a consequential part of the meaning representation for a natural language expression is a set of language-free structures that are not part of the expression in question. To support this claim, we construct a corpus of paraphrase pairs using a system that has a non-linguistic meaning represen- tation decoupled from the linguistic …

Preliminary Perspectives On Information Passing In The Intelligence Community, Jeremy E. Block, Ilana Bookner, Sharon Lynn Chu, R. Jordan Crouser, Donald R. Honeycutt, Rebecca M. Jonas, Abhishek Kulkarni, Yancy Vance Paredes, Eric D. Ragan

Computer Science: Faculty Publications

Analyst sensemaking research typically focuses on individual or small groups conducting intelligence tasks. This has helped understand information retrieval tasks and how people communicate information. As a part of the grand challenge of the Summer Conference on Applied Data Science (SCADS) to build a system that can generate tailored daily reports (TLDR) for intelligence analysts, we conducted a qualitative interview study with analysts to increase understanding of information passing in the intelligence community. While our results are preliminary, we expect that this work will contribute to a better understanding of the information ecosystem of the intelligence community, how institutional dynamics …

Repytah: An Open-Source Python Package For Building Aligned Hierarchies For Sequential Data, Chenhui Jia, Lizette Carpenter, Thu Tran, Amanda Y. Liu, Sasha Yeutseyva, Mariun Tapal, Yingke Wang, Zoie Kexin Zhou, Jordan Moody, Denise Nava, Eleanor Donaher, Lillian Yusha Jiang, Ben Bruncati, Katherine M. Kinnaird

Computer Science: Faculty Publications

We introduce repytah, a Python package that constructs the aligned hierarchies representation that contains all possible structure-based hierarchical decompositions for a finite length piece of sequential data aligned on a common time axis. In particular, this representation–introduced by Kinnaird (2016) with music-based data (like musical recordings or scores) as the primary motivation–is intended for sequential data where repetitions have particular meaning (such as a verse, chorus, motif, or theme). Although the original motivation for the aligned hierarchies representation was finding structure for music-based data streams, there is nothing inherent in the construction of these representations that limits repytah to only …

Metric Ensembles Aid In Explainability: A Case Study With Wikipedia Data, Grant Forbes, R. Jordan Crouser

Computer Science: Faculty Publications

In recent years, as machine learning models have become larger and more complex, it has become both more difficult and more important to be able to explain and interpret the results of those models, both to prevent model errors and to inspire confidence for end users of the model. As such, there has been a significant and growing interest in explainability in recent years as a highly desirable trait for a model to have. Similarly, there has been much recent attention on ensemble methods, which aim to aggregate results from multiple (often simple) models or metrics in order to outperform …

Computing Circuit Polynomials In The Algebraic Rigidity Matroid, Goran Malić, Ileana Streinu

Computer Science: Faculty Publications

We present an algorithm for computing circuit polynomials in the algebraic rigidity matroid A(CM_n) associated to the Cayley-Menger ideal CM_n for n points in 2D. It relies on combinatorial resultants, a new operation on graphs that captures properties of the Sylvester resultant of two polynomials in this ideal. We show that every rigidity circuit has a construction tree from K₄ graphs based on this operation. Our algorithm performs an algebraic elimination guided by such a construction tree, and uses classical resultants, factorization and ideal membership. To highlight its effectiveness, we implemented the algorithm in …

Application Of Traditional Vaccine Development Strategies To Sars-Cov-2, Halie M. Rando, Ronan Lordan, Alexandra J. Lee, Amruta Naik, Nils Wellhausen, Elizabeth Sell, Likhitha Kolla, Anthony Gitter, Casey S. Greene

Computer Science: Faculty Publications

Over the past 150 years, vaccines have revolutionized the relationship between people and disease. During the COVID-19 pandemic, technologies such as mRNA vaccines have received attention due to their novelty and successes. However, more traditional vaccine development platforms have also yielded important tools in the worldwide fight against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). A variety of approaches have been used to develop COVID-19 vaccines that are now authorized for use in countries around the world. In this review, we highlight strategies that focus on the viral capsid and outwards, rather than on the nucleic acids inside. These approaches …

The Coming Of Age Of Nucleic Acid Vaccines During Covid-19, Halie M. Rando, Ronan Lordan, Likhitha Kolla, Elizabeth Sell, Alexandra J. Lee, Nils Wellhausen, Amruta Naik, Jeremy P. Kamil, Anthony Gitter, Casey S. Greene

Computer Science: Faculty Publications

In the 21st century, several emergent viruses have posed a global threat. Each pathogen has emphasized the value of rapid and scalable vaccine development programs. The ongoing severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic has made the importance of such efforts especially clear. New biotechnological advances in vaccinology allow for recent advances that provide only the nucleic acid building blocks of an antigen, eliminating many safety concerns. During the COVID-19 pandemic, these DNA and RNA vaccines have facilitated the development and deployment of vaccines at an unprecedented pace. This success was attributable at least in part to broader shifts …

Coming Out While Going Fast: Queer Conviviality In Speedrunning Live Streams, Johanna Brewer

Computer Science: Faculty Publications

Drawing on ethnographic research of LGBTQIA+ live streaming speedrunners, this article demonstrates how by centering queer perspectives, we can catalyze meaningful social changes for all. Though for most people, beating the original Super Mario Bros. in under five minutes would seem unfathomably difficult, LGBTQIA+ speedrunning live streamers regularly accomplish this superhuman feat, while coming out to an audience of thousands at the same time. For queer and trans folks, broadcasting such a transgressive, transformational form of play is defiant demonstration of vulnerability; one that creates a comfortable space for a community to thrive, by cultivating a culture of queer conviviality. …

Image Schema Decompositions Of The Conceptual Dependency Ingest Primitive: A Study Of Paraphrases, Jamie C. Macbeth, Alexis Kilayko, Zoie Zhao, Sophie Song, Winniw X. Zheng

Computer Science: Faculty Publications

One of the hallmarks of the Schank-Minsky Conceptual Dependency Trans-Frames meaning representation system is that it attempts to express complex meanings by building large and complex conceptual structures using a relatively small number of primitives. Recently comparisons of image schemas with Conceptual Dependency primitives revealed ways of possibly reducing the number of primitives while maintaining the expressiveness of the set—an important research goal because it increases the flexibility and richness of the primitive-decomposed structures in a way that better approximates human cognition. Inspired by this prior work, we employ a paraphrase generation system to explore the replacement of the Conceptual …

An Experiment On The Effects Of Using Color To Visualize Requirements Analysis Tasks: Supplemental Material, Yesugen Baatartogtokh, Irene Foster, Alicia M. Grubb

Computer Science: Faculty Publications

Supplemental material for the paper: "An Experiment on the Effects of using Color to Visualize Requirements Analysis Tasks".
This paper is a scientific evaluation of the effectiveness and usability of EVO. We conduct an experiment to measure any effect of using colors to represent evidence pairs.

A Proposal For Primitive Decomposition Of Spatial Orientation Relationships, Jamie C. Macbeth, Mackie Zhou, Zoie Zhao

Computer Science: Faculty Publications

This short paper continues work on primitive decomposition systems for meaning representation which combine image schemas and conceptual dependency primitive systems. An important thread of this research seeks small abstract sets of conceptual primitives so that decompositions of imagery evoked by language give rise to rich sets of mappings between language and the language-free representations, reflecting the linguistic variation of human language behavior. In this brief paper, we present a proposal for novel primitive decompositions of positions, spatial relationships, and orientations of objects in space in a conceptual representation framework. As an abstract first approximation, we introduce a spatial primitive …

More Human Than Human: Llm-Generated Narratives Outperform Human-Llm Interleaved Narratives, Zoie Zhao, Sophie Song, Bridget Duah, Jamie C. Macbeth, Scott Carter, Monica Van, Nayeli Bravo, Matthew Klenk, Katherine Sieck, Alexandre Filipowicz

Computer Science: Faculty Publications

Narrative story generation has gained emerging interest in the field of large language models. The present paper aims to compare stories generated by an LLM only (non-interleaved) with those generated by interleaving human-generated and LLM-generated text (interleaved). The study’s hypothesis is that interleaved stories would perform better than non-interleaved stories. To verify this hypothesis, we conducted two tests with roughly 500 participants each. Participants were asked to rate stories of each type, including an overall score or preference and four facets—logical soundness, plausibility, understandability, and novelty. Our findings indicate that interleaved stories were in fact less preferred than non-interleaved stories. …

An Experiment On The Effects Of Using Color To Visualize Requirements Analysis Tasks, Yesugen Baatartogtokh, Irene Foster, Alicia M. Grubb

Computer Science: Faculty Publications

Recent approaches have investigated assisting users in making early trade-off decisions when the future evolution of project elements is uncertain. These approaches have demon-strated promise in their analytical capabilities; yet, stakeholders have expressed concerns about the readability of the models and resulting analysis, which builds upon Tropos. Tropos is based on formal semantics enabling automated analysis; however, this creates a problem of interpreting evidence pairs. The aim of our broader research project is to improve the process of model comprehension and decision making by improving how analysts interpret and make decisions. We extend and evaluate a prior approach, called EVO, …