Showing 1–50 of 85 results for author: Anderson, C

The Best Programming Language for Tokenmaxxing: An Investigation of Coding Agent Behavior Across Programming Languages

Authors: Zixuan Wu, Carolyn Jane Anderson, Arjun Guha

Abstract: Although coding agents are now very effective in a variety of programming languages, this paper first shows that the cost (in tokens) can very significantly by programming language. We evaluate five recent models on programming problems in Python, Java, Rust, and OCaml. We carefully control for problem difficulty, and show that there can be stark variation in token consumption that is consistent a… ▽ More Although coding agents are now very effective in a variety of programming languages, this paper first shows that the cost (in tokens) can very significantly by programming language. We evaluate five recent models on programming problems in Python, Java, Rust, and OCaml. We carefully control for problem difficulty, and show that there can be stark variation in token consumption that is consistent across models. To understand why, we analyze both the structure and content of agent trajectories. First, we re-execute every intermediate solution and abstract each trajectory as a sequence of test-outcome vectors, then label the work between successive solutions. This reveals agents repeatedly producing noncompiling solutions in unfamiliar languages and revising solutions that already pass. Second, we analyze trajectory text, finding that agents plan solutions in code comments, distrust the provided tests in favor of inputs they invent, and sidestep unfamiliar target languages by prototyping in Python. Our results show that by-language token efficiency is a metric that should be considered when benchmarking and developing multilingual agents, and, for the tokenmaxxer, a guide to the most expensive language to work in. △ Less

Submitted 24 July, 2026; originally announced July 2026.

Prediction of Challenging Behaviors Associated with Profound Autism in a Classroom Setting Using Wearable Sensors

Authors: Yadhu Kartha, Conor Anderson, Jenny Foster, Theresa Hamlin, Johanna Lantz, Ryan Lay, Juergen Hahn, Gari D. Clifford, Hyeokhyen Kwon

Abstract: Autism Spectrum Disorder (ASD) is characterized by challenges with social interaction and communication and by restricted or repetitive patterns of thought and behavior, with significant variability in presentation. Approximately a quarter of children with ASD are classified as having profound autism, who often exhibit challenging behaviors, such as self-injurious behavior, aggression, elopement,… ▽ More Autism Spectrum Disorder (ASD) is characterized by challenges with social interaction and communication and by restricted or repetitive patterns of thought and behavior, with significant variability in presentation. Approximately a quarter of children with ASD are classified as having profound autism, who often exhibit challenging behaviors, such as self-injurious behavior, aggression, elopement, or pica, that pose serious safety risks and disrupt learning in educational settings. Prior work has applied wearable sensors and machine learning to detect challenging behaviors, but has been largely confined to controlled laboratory environments. This work demonstrates that predicting challenging behavior episodes is feasible in a real-world special education classroom. We collected approximately 110.7 hours of labeled multimodal wearable data comprising accelerometry, electrodermal activity (EDA), and skin temperature from 9 children and young adults aged 10 to 21 years across standard classroom sessions. We fine-tuned state-of-the-art foundation models for multimodal wearable time-series analysis and show that challenging behavior episodes can be predicted up to 10 minutes in advance with an AUC-ROC of 0.78. These results establish a concrete foundation for developing proactive in-class intervention systems that enable teachers to minimize the safety risks of challenging behaviors in special education classrooms △ Less

Submitted 20 May, 2026; v1 submitted 17 May, 2026; originally announced May 2026.

Correctly Rounded Functions For Vector Applications: A Performance Study

Authors: Cristina Anderson, Marius Cornea, Andrey Stepin, Mihai Tudor Panu

Abstract: Following recent interest in correctly rounded math library functions (as currently recommended by the IEEE 754 standard), we have designed several SIMD algorithms for one-input single precision functions and integrated them into our CPU math library; these will form the core of the first correctly rounded vector math library, to be available to users in mid-2026. To take advantage of the cross-pl… ▽ More Following recent interest in correctly rounded math library functions (as currently recommended by the IEEE 754 standard), we have designed several SIMD algorithms for one-input single precision functions and integrated them into our CPU math library; these will form the core of the first correctly rounded vector math library, to be available to users in mid-2026. To take advantage of the cross-platform bitwise reproducibility afforded by correct rounding, we adapted and evaluated a few SIMD implementations on graphics processing units (GPU). In addition, we designed and evaluated proof-of-concept SIMD implementations of two correctly rounded double precision functions. △ Less

Submitted 14 May, 2026; originally announced May 2026.

Comments: 5 pages, 2 figures, 4 tables

Soohak: A Mathematician-Curated Benchmark for Evaluating Research-level Math Capabilities of LLMs

Authors: Guijin Son, Seungone Kim, Catherine Arnett, Hyunwoo Ko, Hyein Lee, Hyeonah Kang, Jiang Longxi, Jin Yun, JungYup Lee, Kyungmin Lee, Sam Yoosuk Kim, Sang Park, Seunghyeok Hong, SeungJae Lee, Seungyeop Yi, Shinae Shin, SunHye Bok, Sunyoung Shin, Yonghoon Ji, Youngtaek Kim, Hanearl Jung, Akari Asai, Graham Neubig, Sean Welleck, Youngjae Yu , et al. (51 additional authors not shown)

Abstract: Following the recent achievement of gold-medal performance on the IMO by frontier LLMs, the community is searching for the next meaningful and challenging target for measuring LLM reasoning. Whereas olympiad-style problems measure step-by-step reasoning alone, research-level problems use such reasoning to advance the frontier of mathematical knowledge itself, emerging as a compelling alternative.… ▽ More Following the recent achievement of gold-medal performance on the IMO by frontier LLMs, the community is searching for the next meaningful and challenging target for measuring LLM reasoning. Whereas olympiad-style problems measure step-by-step reasoning alone, research-level problems use such reasoning to advance the frontier of mathematical knowledge itself, emerging as a compelling alternative. Yet research-level math benchmarks remain scarce because such problems are difficult to source (e.g., Riemann Bench and FrontierMath-Tier 4 contain 25 and 50 problems, respectively). To support reliable evaluation of next-generation frontier models, we introduce Soohak, a 439-problem benchmark newly authored from scratch by 64 mathematicians. Soohak comprises two subsets. On the Challenge subset, frontier models including Gemini-3-Pro, GPT-5, and Claude-Opus-4.5 reach 30.4%, 26.4%, and 10.4% respectively, leaving substantial headroom, while leading open-weight models such as Qwen3-235B, GPT-OSS-120B, and Kimi-2.5 remain below 15%. Notably, beyond standard problem solving, Soohak introduces a refusal subset that probes a capability intrinsic to research mathematics: recognizing ill-posed problems and pausing rather than producing confident but unjustified answers. On this subset, no model exceeds 50%, identifying refusal as a new optimization target that current models do not directly address. To prevent contamination, the dataset will be publicly released in late 2026, with model evaluations available upon request in the interim. △ Less

Submitted 19 May, 2026; v1 submitted 9 May, 2026; originally announced May 2026.

Comments: Under review, For questions or model-evaluation requests, contact $guijin.son@snu.ac.kr$

Open-access model for detecting openly dumped dispersed municipal solid waste from crowdsourced UAV imagery in Sub-Saharan Africa

Authors: Steffen Knoblauch, Ram Kumar Muthusamy, Luis M. A. Bettencourt, Costas Velis, Pierre Chrzanowski, Edward Charles Anderson, Pete Masters, Innocent Maholi, Antonio Inguane, Levi Szamek, Alexander Zipf

Abstract: Managing municipal solid waste in rapidly urbanizing Sub-Saharan Africa remains challenging due to dispersed informal dumping and limited high-resolution datasets for spatial monitoring. We present an open-access deep learning model for automated detection of openly dumped dispersed solid waste via crowdsourced UAV imagery, trained and evaluated across 29 regions in 10 countries, encompassing dive… ▽ More Managing municipal solid waste in rapidly urbanizing Sub-Saharan Africa remains challenging due to dispersed informal dumping and limited high-resolution datasets for spatial monitoring. We present an open-access deep learning model for automated detection of openly dumped dispersed solid waste via crowdsourced UAV imagery, trained and evaluated across 29 regions in 10 countries, encompassing diverse environmental contexts. A deep learning model trained on manually annotated image tiles achieved excellent performance in detecting openly dumped dispersed solid waste across all study regions. Predicted distributions reveal heterogeneous accumulation patterns, ranging from localized hotspots - often along waterways, where waste can exacerbate flood and public health risks - to more dispersed litter across urban areas. Waste accumulation is most strongly associated with population density and indicators of lack of local infrastructure access, whereas its relationship with broader measures of regional development is weaker, highlighting the importance of fine-scale data for understanding localized waste dynamics. By releasing the model, this study provides a ready-to-use tool for UAV imagery collected by municipalities and local mapping communities, enabling openly dumped dispersed solid waste monitoring without extensive technical expertise. This approach empowers local practitioners to convert UAV imagery into actionable insights, supporting targeted interventions and improved municipal solid waste management across Sub-Saharan Africa. △ Less

Submitted 4 May, 2026; originally announced May 2026.

Understanding and Mitigating Dataset Corruption in LLM Steering

Authors: Cullen Anderson, Narmeen Oozeer, Foad Namjoo, Remy Ogasawara, Amirali Abdullah, Jeff M. Phillips

Abstract: Contrastive steering has been shown as a simple and effective method to adjust the generative behavior of LLMs at inference time. It uses examples of prompt responses with and without a trait to identify a direction in an intermediate activation layer, and then shifts activations in this 1-dimensional subspace. However, despite its growing use in AI safety applications, the robustness of contrasti… ▽ More Contrastive steering has been shown as a simple and effective method to adjust the generative behavior of LLMs at inference time. It uses examples of prompt responses with and without a trait to identify a direction in an intermediate activation layer, and then shifts activations in this 1-dimensional subspace. However, despite its growing use in AI safety applications, the robustness of contrastive steering to noisy or adversarial data corruption is poorly understood. We initiate a study of the robustness of this process with respect to corruption of the dataset of examples used to train the steering direction. Our first observation is that contrastive steering is quite robust to a moderate amount of corruption, but unwanted side effects can be clearly and maliciously manifested when a non-trivial fraction of the training data is altered. Second, we analyze the geometry of various types of corruption, and identify some safeguards. Notably, a key step in learning the steering direction involves high-dimensional mean computation, and we show that replacing this step with a recently developed robust mean estimator often mitigates most of the unwanted effects of malicious corruption. △ Less

Submitted 3 March, 2026; originally announced March 2026.

Fauna Sprout: A lightweight, approachable, developer-ready humanoid robot

Authors: Fauna Robotics, :, Diego Aldarondo, Ana Pervan, Daniel Corbalan, Dave Petrillo, Bolun Dai, Aadhithya Iyer, Nina Mortensen, Erik Pearson, Sridhar Pandian Arunachalam, Emma Reznick, David Weis, Jacob Davison, Samuel Patterson, Tess Carella, Michael Suguitan, David Ye, Oswaldo Ferro, Nilesh Suriyarachchi, Spencer Ling, Erik Su, Daniel Giebisch, Peter Traver, Sam Fonseca , et al. (26 additional authors not shown)

Abstract: Recent advances in learned control, large-scale simulation, and generative models have accelerated progress toward general-purpose robotic controllers, yet the field still lacks platforms suitable for safe, expressive, long-term deployment in human environments. Most existing humanoids are either closed industrial systems or academic prototypes that are difficult to deploy and operate around peopl… ▽ More Recent advances in learned control, large-scale simulation, and generative models have accelerated progress toward general-purpose robotic controllers, yet the field still lacks platforms suitable for safe, expressive, long-term deployment in human environments. Most existing humanoids are either closed industrial systems or academic prototypes that are difficult to deploy and operate around people, limiting progress in robotics. We introduce Sprout, a developer platform designed to address these limitations through an emphasis on safety, expressivity, and developer accessibility. Sprout adopts a lightweight form factor with compliant control, limited joint torques, and soft exteriors to support safe operation in shared human spaces. The platform integrates whole-body control, manipulation with integrated grippers, and virtual-reality-based teleoperation within a unified hardware-software stack. An expressive head further enables social interaction -- a domain that remains underexplored on most utilitarian humanoids. By lowering physical and technical barriers to deployment, Sprout expands access to capable humanoid platforms and provides a practical basis for developing embodied intelligence in real human environments. △ Less

Submitted 26 January, 2026; originally announced January 2026.

The Geometry of Thought: How Scale Restructures Reasoning In Large Language Models

Authors: Samuel Cyrenius Anderson

Abstract: Scale does not uniformly improve reasoning - it restructures it. Analyzing 25,000+ chain-of-thought trajectories across four domains (Law, Science, Code, Math) and two scales (8B, 70B parameters), we discover that neural scaling laws trigger domain-specific phase transitions rather than uniform capability gains. Legal reasoning undergoes Crystallization: 45% collapse in representational dimensiona… ▽ More Scale does not uniformly improve reasoning - it restructures it. Analyzing 25,000+ chain-of-thought trajectories across four domains (Law, Science, Code, Math) and two scales (8B, 70B parameters), we discover that neural scaling laws trigger domain-specific phase transitions rather than uniform capability gains. Legal reasoning undergoes Crystallization: 45% collapse in representational dimensionality (d95: 501 -> 274), 31% increase in trajectory alignment, and 10x manifold untangling. Scientific and mathematical reasoning remain Liquid - geometrically invariant despite 9x parameter increase. Code reasoning forms a discrete Lattice of strategic modes (silhouette: 0.13 -> 0.42). This geometry predicts learnability. We introduce Neural Reasoning Operators - learned mappings from initial to terminal hidden states. In crystalline legal reasoning, our operator achieves 63.6% accuracy on held-out tasks via probe decoding, predicting reasoning endpoints without traversing intermediate states. We further identify a universal oscillatory signature (coherence ~ -0.4) invariant across domains and scales, suggesting attention and feedforward layers drive reasoning through opposing dynamics. These findings establish that the cost of thought is determined not by task difficulty but by manifold geometry - offering a blueprint for inference acceleration where topology permits. △ Less

Submitted 30 March, 2026; v1 submitted 19 January, 2026; originally announced January 2026.

Comments: The theoretical framework has been shown to be wrong and should not be followed for future research direction

ACM Class: I.2.6; I.2.7

Agnostics: Learning to Code in Any Programming Language via Reinforcement with a Universal Learning Environment

Authors: Aleksander Boruch-Gruszecki, Yangtian Zi, Zixuan Wu, Tejas Oberoi, Carolyn Jane Anderson, Joydeep Biswas, Arjun Guha

Abstract: Large language models (LLMs) already excel at writing code in high-resource languages such as Python and JavaScript, yet stumble on low-resource languages that remain essential to science and engineering. Besides the obvious shortage of pre-training data, post-training itself is a bottleneck: every new language seems to require new datasets, test harnesses, and reinforcement-learning (RL) infrastr… ▽ More Large language models (LLMs) already excel at writing code in high-resource languages such as Python and JavaScript, yet stumble on low-resource languages that remain essential to science and engineering. Besides the obvious shortage of pre-training data, post-training itself is a bottleneck: every new language seems to require new datasets, test harnesses, and reinforcement-learning (RL) infrastructure. We introduce Agnostics, a language-agnostic post-training pipeline that eliminates this per-language engineering. The key idea is to judge code solely by its externally observable behavior, so a single verifier can test solutions written in any language. Concretely, we (i) use an LLM to rewrite existing unit-test datasets into an I/O format, (ii) supply a short configuration that tells the verifier how to compile and run a target language, and (iii) apply reinforcement learning with verifiable rewards (RLVR) in a robust code execution environment. Applied to five low-resource languages--Lua, Julia, R, OCaml, and Fortran--Agnostics (1) improves Qwen-3 4B to performance that rivals other 16B-70B open-weight models; (2) scales cleanly to larger and diverse model families (Qwen-3 8B, DeepSeek Coder 6.7B Instruct, Phi 4 Mini); and (3) for ${\le} 16$B parameter models, sets new state-of-the-art pass@1 results on MultiPL-E and a new multi-language version of LiveCodeBench that we introduce. We release the language-agnostic training datasets (Ag-MBPP-X, Ag-Codeforces-X, Ag-LiveCodeBench-X), training code, and ready-to-use configurations, making RL post-training in any programming language as simple as editing a short YAML file. △ Less

Submitted 23 March, 2026; v1 submitted 6 August, 2025; originally announced August 2025.

Comments: 30 pages, 19 figures. Accepted at ICLR 2026. For data, code, artifacts, see https://agnostics.abgru.me

Consistency in Language Models: Current Landscape, Challenges, and Future Directions

Authors: Jekaterina Novikova, Carol Anderson, Borhane Blili-Hamelin, Domenic Rosati, Subhabrata Majumdar

Abstract: The hallmark of effective language use lies in consistency: expressing similar meanings in similar contexts and avoiding contradictions. While human communication naturally demonstrates this principle, state-of-the-art language models (LMs) struggle to maintain reliable consistency across task- and domain-specific applications. Here we examine the landscape of consistency research in LMs, analyze… ▽ More The hallmark of effective language use lies in consistency: expressing similar meanings in similar contexts and avoiding contradictions. While human communication naturally demonstrates this principle, state-of-the-art language models (LMs) struggle to maintain reliable consistency across task- and domain-specific applications. Here we examine the landscape of consistency research in LMs, analyze current approaches to measure aspects of consistency, and identify critical research gaps. Our findings point to an urgent need for quality benchmarks to measure and interdisciplinary approaches to ensure consistency while preserving utility. △ Less

Submitted 11 July, 2025; v1 submitted 30 April, 2025; originally announced May 2025.

Comments: Accepted in ICML 2025 Workshop on Reliable and Responsible Foundation Models

"I Would Have Written My Code Differently'': Beginners Struggle to Understand LLM-Generated Code

Authors: Yangtian Zi, Luisa Li, Arjun Guha, Carolyn Jane Anderson, Molly Q Feldman

Abstract: Large language models (LLMs) are being increasingly adopted for programming work. Prior work shows that while LLMs accelerate task completion for professional programmers, beginning programmers struggle to prompt models effectively. However, prompting is just half of the code generation process -- when code is generated, it must be read, evaluated, and integrated (or rejected). How accessible are… ▽ More Large language models (LLMs) are being increasingly adopted for programming work. Prior work shows that while LLMs accelerate task completion for professional programmers, beginning programmers struggle to prompt models effectively. However, prompting is just half of the code generation process -- when code is generated, it must be read, evaluated, and integrated (or rejected). How accessible are these tasks for beginning programmers? This paper measures how well beginners comprehend LLM-generated code and explores the challenges students face in judging code correctness. We compare how well students understand natural language descriptions of functions and LLM-generated implementations, studying 32 CS1 students on 160 task instances. Our results show a low per-task success rate of 32.5\%, with indiscriminate struggles across demographic populations. Key challenges include barriers for non-native English speakers, unfamiliarity with Python syntax, and automation bias. Our findings highlight the barrier that code comprehension presents to beginning programmers seeking to write code with LLMs. △ Less

Submitted 26 April, 2025; originally announced April 2025.

Comments: To appear in 33rd ACM International Conference on the Foundations of Software Engineering (FSE Companion '25), June 23-28, 2025, Trondheim, Norway

Generalization is not a universal guarantee: Estimating similarity to training data with an ensemble out-of-distribution metric

Authors: W. Max Schreyer, Christopher Anderson, Reid F. Thompson

Abstract: Failure of machine learning models to generalize to new data is a core problem limiting the reliability of AI systems, partly due to the lack of simple and robust methods for comparing new data to the original training dataset. We propose a standardized approach for assessing data similarity in a model-agnostic manner by constructing a supervised autoencoder for generalizability estimation (SAGE).… ▽ More Failure of machine learning models to generalize to new data is a core problem limiting the reliability of AI systems, partly due to the lack of simple and robust methods for comparing new data to the original training dataset. We propose a standardized approach for assessing data similarity in a model-agnostic manner by constructing a supervised autoencoder for generalizability estimation (SAGE). We compare points in a low-dimensional embedded latent space, defining empirical probability measures for k-Nearest Neighbors (kNN) distance, reconstruction of inputs and task-based performance. As proof of concept for classification tasks, we use MNIST and CIFAR-10 to demonstrate how an ensemble output probability score can separate deformed images from a mixture of typical test examples, and how this SAGE score is robust to transformations of increasing severity. As further proof of concept, we extend this approach to a regression task using non-imaging data (UCI Abalone). In all cases, we show that out-of-the-box model performance increases after SAGE score filtering, even when applied to data from the model's own training and test datasets. Our out-of-distribution scoring method can be introduced during several steps of model construction and assessment, leading to future improvements in responsible deep learning implementation. △ Less

Submitted 25 February, 2025; v1 submitted 22 February, 2025; originally announced February 2025.

Comments: 10 pages, 5 figures

Robust High-Dimensional Mean Estimation With Low Data Size, an Empirical Study

Authors: Cullen Anderson, Jeff M. Phillips

Abstract: Robust statistics aims to compute quantities to represent data where a fraction of it may be arbitrarily corrupted. The most essential statistic is the mean, and in recent years, there has been a flurry of theoretical advancement for efficiently estimating the mean in high dimensions on corrupted data. While several algorithms have been proposed that achieve near-optimal error, they all rely on la… ▽ More Robust statistics aims to compute quantities to represent data where a fraction of it may be arbitrarily corrupted. The most essential statistic is the mean, and in recent years, there has been a flurry of theoretical advancement for efficiently estimating the mean in high dimensions on corrupted data. While several algorithms have been proposed that achieve near-optimal error, they all rely on large data size requirements as a function of dimension. In this paper, we perform an extensive experimentation over various mean estimation techniques where data size might not meet this requirement due to the high-dimensional setting. △ Less

Submitted 16 February, 2025; originally announced February 2025.

Journal ref: Transactions on Machine Learning Research (TMLR), February 2025, ISSN: 2835-8856

ReasoningWeekly: A General Knowledge and Verbal Reasoning Challenge for Large Language Models

Authors: Zixuan Wu, Francesca Lucchetti, Aleksander Boruch-Gruszecki, Jingmiao Zhao, Carolyn Jane Anderson, Joydeep Biswas, Federico Cassano, Arjun Guha

Abstract: Existing benchmarks for frontier models often test specialized, "PhD-level" knowledge that is difficult for non-experts to grasp. In contrast, we present a benchmark with 613 problems based on the NPR Sunday Puzzle Challenge that requires only general knowledge. Our benchmark is challenging for both humans and models; however correct solutions are easy to verify, and models' mistakes are easy to s… ▽ More Existing benchmarks for frontier models often test specialized, "PhD-level" knowledge that is difficult for non-experts to grasp. In contrast, we present a benchmark with 613 problems based on the NPR Sunday Puzzle Challenge that requires only general knowledge. Our benchmark is challenging for both humans and models; however correct solutions are easy to verify, and models' mistakes are easy to spot. As LLMs are more widely deployed in society, we believe it is useful to develop benchmarks for frontier models that humans can understand without the need for deep domain expertise. Our work reveals capability gaps that are not evident in existing benchmarks: OpenAI o1 significantly outperforms other reasoning models on our benchmark, despite being on par with other models when tested on benchmarks that test specialized knowledge. Furthermore, our analysis of reasoning outputs uncovers new kinds of failures. DeepSeek R1, for instance, often concedes with "I give up" before providing an answer that it knows is wrong. R1 can also be remarkably "uncertain" in its output and in rare cases, it does not "finish thinking," which suggests the need for techniques to ``wrap up'' before the context window limit is reached. We also quantify the effectiveness of reasoning longer to identify the point beyond which more reasoning is unlikely to improve accuracy on our benchmark. △ Less

Submitted 26 November, 2025; v1 submitted 3 February, 2025; originally announced February 2025.

A generalizable 3D framework and model for self-supervised learning in medical imaging

Authors: Tony Xu, Sepehr Hosseini, Chris Anderson, Anthony Rinaldi, Rahul G. Krishnan, Anne L. Martel, Maged Goubran

Abstract: Current self-supervised learning methods for 3D medical imaging rely on simple pretext formulations and organ- or modality-specific datasets, limiting their generalizability and scalability. We present 3DINO, a cutting-edge SSL method adapted to 3D datasets, and use it to pretrain 3DINO-ViT: a general-purpose medical imaging model, on an exceptionally large, multimodal, and multi-organ dataset of… ▽ More Current self-supervised learning methods for 3D medical imaging rely on simple pretext formulations and organ- or modality-specific datasets, limiting their generalizability and scalability. We present 3DINO, a cutting-edge SSL method adapted to 3D datasets, and use it to pretrain 3DINO-ViT: a general-purpose medical imaging model, on an exceptionally large, multimodal, and multi-organ dataset of ~100,000 3D medical imaging scans from over 10 organs. We validate 3DINO-ViT using extensive experiments on numerous medical imaging segmentation and classification tasks. Our results demonstrate that 3DINO-ViT generalizes across modalities and organs, including out-of-distribution tasks and datasets, outperforming state-of-the-art methods on the majority of evaluation metrics and labeled dataset sizes. Our 3DINO framework and 3DINO-ViT will be made available to enable research on 3D foundation models or further finetuning for a wide range of medical imaging applications. △ Less

Submitted 8 June, 2026; v1 submitted 20 January, 2025; originally announced January 2025.

Comments: Published in npj Digital Medicine

Substance Beats Style: Why Beginning Students Fail to Code with LLMs

Authors: Francesca Lucchetti, Zixuan Wu, Arjun Guha, Molly Q Feldman, Carolyn Jane Anderson

Abstract: Although LLMs are increasing the productivity of professional programmers, existing work shows that beginners struggle to prompt LLMs to solve text-to-code tasks. Why is this the case? This paper explores two competing hypotheses about the cause of student-LLM miscommunication: (1) students simply lack the technical vocabulary needed to write good prompts, and (2) students do not understand the ex… ▽ More Although LLMs are increasing the productivity of professional programmers, existing work shows that beginners struggle to prompt LLMs to solve text-to-code tasks. Why is this the case? This paper explores two competing hypotheses about the cause of student-LLM miscommunication: (1) students simply lack the technical vocabulary needed to write good prompts, and (2) students do not understand the extent of information that LLMs need to solve code generation tasks. We study (1) with a causal intervention experiment on technical vocabulary and (2) by analyzing graphs that abstract how students edit prompts and the different failures that they encounter. We find that substance beats style: a poor grasp of technical vocabulary is merely correlated with prompt failure; that the information content of prompts predicts success; that students get stuck making trivial edits; and more. Our findings have implications for the use of LLMs in programming education, and for efforts to make computing more accessible with LLMs. △ Less

Submitted 15 October, 2024; originally announced October 2024.

Evaluating Computational Representations of Character: An Austen Character Similarity Benchmark

Authors: Funing Yang, Carolyn Jane Anderson

Abstract: Several systems have been developed to extract information about characters to aid computational analysis of English literature. We propose character similarity grouping as a holistic evaluation task for these pipelines. We present AustenAlike, a benchmark suite of character similarities in Jane Austen's novels. Our benchmark draws on three notions of character similarity: a structurally defined n… ▽ More Several systems have been developed to extract information about characters to aid computational analysis of English literature. We propose character similarity grouping as a holistic evaluation task for these pipelines. We present AustenAlike, a benchmark suite of character similarities in Jane Austen's novels. Our benchmark draws on three notions of character similarity: a structurally defined notion of similarity; a socially defined notion of similarity; and an expert defined set extracted from literary criticism. We use AustenAlike to evaluate character features extracted using two pipelines, BookNLP and FanfictionNLP. We build character representations from four kinds of features and compare them to the three AustenAlike benchmarks and to GPT-4 similarity rankings. We find that though computational representations capture some broad similarities based on shared social and narrative roles, the expert pairings in our third benchmark are challenging for all systems, highlighting the subtler aspects of similarity noted by human readers. △ Less

Submitted 28 August, 2024; originally announced August 2024.

GlyphPattern: An Abstract Pattern Recognition Benchmark for Vision-Language Models

Authors: Zixuan Wu, Yoolim Kim, Carolyn Jane Anderson

Abstract: Vision-Language Models (VLMs) building upon the foundation of powerful large language models have made rapid progress in reasoning across visual and textual data. While VLMs perform well on vision tasks that they are trained on, our results highlight key challenges in abstract pattern recognition. We present GlyphPattern, a 954 item dataset that pairs 318 human-written descriptions of visual patte… ▽ More Vision-Language Models (VLMs) building upon the foundation of powerful large language models have made rapid progress in reasoning across visual and textual data. While VLMs perform well on vision tasks that they are trained on, our results highlight key challenges in abstract pattern recognition. We present GlyphPattern, a 954 item dataset that pairs 318 human-written descriptions of visual patterns from 40 writing systems with three visual presentation styles. GlyphPattern evaluates abstract pattern recognition in VLMs, requiring models to understand and judge natural language descriptions of visual patterns. GlyphPattern patterns are drawn from a large-scale cognitive science investigation of human writing systems; as a result, they are rich in spatial reference and compositionality. Our experiments show that GlyphPattern is challenging for state-of-the-art VLMs (GPT-4o achieves only 55% accuracy), with marginal gains from few-shot prompting. Our detailed error analysis reveals challenges at multiple levels, including visual processing, natural language understanding, and pattern generalization. △ Less

Submitted 24 June, 2025; v1 submitted 11 August, 2024; originally announced August 2024.

Journal ref: Findings of the ACL 2025

Explainable Artificial Intelligence for Quantifying Interfering and High-Risk Behaviors in Autism Spectrum Disorder in a Real-World Classroom Environment Using Privacy-Preserving Video Analysis

Authors: Barun Das, Conor Anderson, Tania Villavicencio, Johanna Lantz, Jenny Foster, Theresa Hamlin, Ali Bahrami Rad, Gari D. Clifford, Hyeokhyen Kwon

Abstract: Rapid identification and accurate documentation of interfering and high-risk behaviors in ASD, such as aggression, self-injury, disruption, and restricted repetitive behaviors, are important in daily classroom environments for tracking intervention effectiveness and allocating appropriate resources to manage care needs. However, having a staff dedicated solely to observing is costly and uncommon i… ▽ More Rapid identification and accurate documentation of interfering and high-risk behaviors in ASD, such as aggression, self-injury, disruption, and restricted repetitive behaviors, are important in daily classroom environments for tracking intervention effectiveness and allocating appropriate resources to manage care needs. However, having a staff dedicated solely to observing is costly and uncommon in most educational settings. Recently, multiple research studies have explored developing automated, continuous, and objective tools using machine learning models to quantify behaviors in ASD. However, the majority of the work was conducted under a controlled environment and has not been validated for real-world conditions. In this work, we demonstrate that the latest advances in video-based group activity recognition techniques can quantify behaviors in ASD in real-world activities in classroom environments while preserving privacy. Our explainable model could detect the episode of problem behaviors with a 77% F1-score and capture distinctive behavior features in different types of behaviors in ASD. To the best of our knowledge, this is the first work that shows the promise of objectively quantifying behaviors in ASD in a real-world environment, which is an important step toward the development of a practical tool that can ease the burden of data collection for classroom staff. △ Less

Submitted 31 July, 2024; originally announced July 2024.

Distributed Instruments for Planetary Surface Science: Scientific Opportunities and Technology Feasibility

Authors: Federico Rossi, Robert C. Anderson, Saptarshi Bandyopadhyay, Erik Brandon, Ashish Goel, Joshua Vander Hook, Michael Mischna, Michaela Villarreal, Mark Wronkiewicz

Abstract: In this paper, we assess the scientific promise and technology feasibility of distributed instruments for planetary science. A distributed instrument is an instrument designed to collect spatially and temporally correlated data from multiple networked, geographically distributed point sensors. Distributed instruments are ubiquitous in Earth science, where they are routinely employed for weather an… ▽ More In this paper, we assess the scientific promise and technology feasibility of distributed instruments for planetary science. A distributed instrument is an instrument designed to collect spatially and temporally correlated data from multiple networked, geographically distributed point sensors. Distributed instruments are ubiquitous in Earth science, where they are routinely employed for weather and climate science, seismic studies and resource prospecting, and detection of industrial emissions. However, to date, their adoption in planetary surface science has been minimal. It is natural to ask whether this lack of adoption is driven by low potential to address high-priority questions in planetary science; immature technology; or both. To address this question, we survey high-priority planetary science questions that are uniquely well-suited to distributed instruments. We identify four areas of research where distributed instruments hold promise to unlock answers that are largely inaccessible to monolithic sensors, namely, weather and climate studies of Mars; localization of seismic events on rocky and icy bodies; localization of trace gas emissions, primarily on Mars; and magnetometry studies of internal composition. Next, we survey enabling technologies for distributed sensors and assess their maturity. We identify sensor placement (including descent and landing on planetary surfaces), power, and instrument autonomy as three key areas requiring further investment to enable future distributed instruments. Overall, this work shows that distributed instruments hold great promise for planetary science, and paves the way for follow-on studies of future distributed instruments for Solar System in-situ science. △ Less

Submitted 1 July, 2024; originally announced July 2024.

SRViT: Vision Transformers for Estimating Radar Reflectivity from Satellite Observations at Scale

Authors: Jason Stock, Kyle Hilburn, Imme Ebert-Uphoff, Charles Anderson

Abstract: We introduce a transformer-based neural network to generate high-resolution (3km) synthetic radar reflectivity fields at scale from geostationary satellite imagery. This work aims to enhance short-term convective-scale forecasts of high-impact weather events and aid in data assimilation for numerical weather prediction over the United States. Compared to convolutional approaches, which have limite… ▽ More We introduce a transformer-based neural network to generate high-resolution (3km) synthetic radar reflectivity fields at scale from geostationary satellite imagery. This work aims to enhance short-term convective-scale forecasts of high-impact weather events and aid in data assimilation for numerical weather prediction over the United States. Compared to convolutional approaches, which have limited receptive fields, our results show improved sharpness and higher accuracy across various composite reflectivity thresholds. Additional case studies over specific atmospheric phenomena support our quantitative findings, while a novel attribution method is introduced to guide domain experts in understanding model outputs. △ Less

Submitted 28 June, 2024; v1 submitted 20 June, 2024; originally announced June 2024.

Comments: Published as a workshop paper at "Machine Learning for Earth System Modeling", ICML 2024; added acknowledgements and github link

Self-training superconducting neuromorphic circuits using reinforcement learning rules

Authors: M. L. Schneider, E. M. Jué, M. R. Pufall, K. Segall, C. W. Anderson

Abstract: Reinforcement learning algorithms are used in a wide range of applications, from gaming and robotics to autonomous vehicles. In this paper we describe a set of reinforcement learning-based local weight update rules and their implementation in superconducting hardware. Using SPICE circuit simulations, we implement a small-scale neural network with a learning time of order one nanosecond. This netwo… ▽ More Reinforcement learning algorithms are used in a wide range of applications, from gaming and robotics to autonomous vehicles. In this paper we describe a set of reinforcement learning-based local weight update rules and their implementation in superconducting hardware. Using SPICE circuit simulations, we implement a small-scale neural network with a learning time of order one nanosecond. This network can be trained to learn new functions simply by changing the target output for a given set of inputs, without the need for any external adjustments to the network. In this implementation the weights are adjusted based on the current state of the overall network response and locally stored information about the previous action. This removes the need to program explicit weight values in these networks, which is one of the primary challenges that analog hardware implementations of neural networks face. The adjustment of weights is based on a global reinforcement signal that obviates the need for circuitry to back-propagate errors. △ Less

Submitted 29 April, 2024; originally announced April 2024.

Comments: 15 pages, 6 figures

Journal ref: npj Unconventional Computing. 2, 5 (2025)

StarCoder 2 and The Stack v2: The Next Generation

Authors: Anton Lozhkov, Raymond Li, Loubna Ben Allal, Federico Cassano, Joel Lamy-Poirier, Nouamane Tazi, Ao Tang, Dmytro Pykhtar, Jiawei Liu, Yuxiang Wei, Tianyang Liu, Max Tian, Denis Kocetkov, Arthur Zucker, Younes Belkada, Zijian Wang, Qian Liu, Dmitry Abulkhanov, Indraneil Paul, Zhuang Li, Wen-Ding Li, Megan Risdal, Jia Li, Jian Zhu, Terry Yue Zhuo , et al. (41 additional authors not shown)

Abstract: The BigCode project, an open-scientific collaboration focused on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder2. In partnership with Software Heritage (SWH), we build The Stack v2 on top of the digital commons of their source code archive. Alongside the SWH repositories spanning 619 programming languages, we carefully select other high-quality data… ▽ More The BigCode project, an open-scientific collaboration focused on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder2. In partnership with Software Heritage (SWH), we build The Stack v2 on top of the digital commons of their source code archive. Alongside the SWH repositories spanning 619 programming languages, we carefully select other high-quality data sources, such as GitHub pull requests, Kaggle notebooks, and code documentation. This results in a training set that is 4x larger than the first StarCoder dataset. We train StarCoder2 models with 3B, 7B, and 15B parameters on 3.3 to 4.3 trillion tokens and thoroughly evaluate them on a comprehensive set of Code LLM benchmarks. We find that our small model, StarCoder2-3B, outperforms other Code LLMs of similar size on most benchmarks, and also outperforms StarCoderBase-15B. Our large model, StarCoder2- 15B, significantly outperforms other models of comparable size. In addition, it matches or outperforms CodeLlama-34B, a model more than twice its size. Although DeepSeekCoder- 33B is the best-performing model at code completion for high-resource languages, we find that StarCoder2-15B outperforms it on math and code reasoning benchmarks, as well as several low-resource languages. We make the model weights available under an OpenRAIL license and ensure full transparency regarding the training data by releasing the SoftWare Heritage persistent IDentifiers (SWHIDs) of the source code data. △ Less

Submitted 29 February, 2024; originally announced February 2024.

Simulation-Enhanced Data Augmentation for Machine Learning Pathloss Prediction

Authors: Ahmed P. Mohamed, Byunghyun Lee, Yaguang Zhang, Max Hollingsworth, C. Robert Anderson, James V. Krogmeier, David J. Love

Abstract: Machine learning (ML) offers a promising solution to pathloss prediction. However, its effectiveness can be degraded by the limited availability of data. To alleviate these challenges, this paper introduces a novel simulation-enhanced data augmentation method for ML pathloss prediction. Our method integrates synthetic data generated from a cellular coverage simulator and independently collected re… ▽ More Machine learning (ML) offers a promising solution to pathloss prediction. However, its effectiveness can be degraded by the limited availability of data. To alleviate these challenges, this paper introduces a novel simulation-enhanced data augmentation method for ML pathloss prediction. Our method integrates synthetic data generated from a cellular coverage simulator and independently collected real-world datasets. These datasets were collected through an extensive measurement campaign in different environments, including farms, hilly terrains, and residential areas. This comprehensive data collection provides vital ground truth for model training. A set of channel features was engineered, including geographical attributes derived from LiDAR datasets. These features were then used to train our prediction model, incorporating the highly efficient and robust gradient boosting ML algorithm, CatBoost. The integration of synthetic data, as demonstrated in our study, significantly improves the generalizability of the model in different environments, achieving a remarkable improvement of approximately 12dB in terms of mean absolute error for the best-case scenario. Moreover, our analysis reveals that even a small fraction of measurements added to the simulation training set, with proper data balance, can significantly enhance the model's performance. △ Less

Submitted 5 February, 2024; v1 submitted 2 February, 2024; originally announced February 2024.

Comments: 6 pages, 5 figures, Accepted at ICC 2024

How Beginning Programmers and Code LLMs (Mis)read Each Other

Authors: Sydney Nguyen, Hannah McLean Babe, Yangtian Zi, Arjun Guha, Carolyn Jane Anderson, Molly Q Feldman

Abstract: Generative AI models, specifically large language models (LLMs), have made strides towards the long-standing goal of text-to-code generation. This progress has invited numerous studies of user interaction. However, less is known about the struggles and strategies of non-experts, for whom each step of the text-to-code problem presents challenges: describing their intent in natural language, evaluat… ▽ More Generative AI models, specifically large language models (LLMs), have made strides towards the long-standing goal of text-to-code generation. This progress has invited numerous studies of user interaction. However, less is known about the struggles and strategies of non-experts, for whom each step of the text-to-code problem presents challenges: describing their intent in natural language, evaluating the correctness of generated code, and editing prompts when the generated code is incorrect. This paper presents a large-scale controlled study of how 120 beginning coders across three academic institutions approach writing and editing prompts. A novel experimental design allows us to target specific steps in the text-to-code process and reveals that beginners struggle with writing and editing prompts, even for problems at their skill level and when correctness is automatically determined. Our mixed-methods evaluation provides insight into student processes and perceptions with key implications for non-expert Code LLM use within and outside of education. △ Less

Submitted 7 July, 2024; v1 submitted 26 January, 2024; originally announced January 2024.

Comments: Published in CHI 2024

Segmenting Messy Text: Detecting Boundaries in Text Derived from Historical Newspaper Images

Authors: Carol Anderson, Phil Crone

Abstract: Text segmentation, the task of dividing a document into sections, is often a prerequisite for performing additional natural language processing tasks. Existing text segmentation methods have typically been developed and tested using clean, narrative-style text with segments containing distinct topics. Here we consider a challenging text segmentation task: dividing newspaper marriage announcement l… ▽ More Text segmentation, the task of dividing a document into sections, is often a prerequisite for performing additional natural language processing tasks. Existing text segmentation methods have typically been developed and tested using clean, narrative-style text with segments containing distinct topics. Here we consider a challenging text segmentation task: dividing newspaper marriage announcement lists into units of one announcement each. In many cases the information is not structured into sentences, and adjacent segments are not topically distinct from each other. In addition, the text of the announcements, which is derived from images of historical newspapers via optical character recognition, contains many typographical errors. As a result, these announcements are not amenable to segmentation with existing techniques. We present a novel deep learning-based model for segmenting such text and show that it significantly outperforms an existing state-of-the-art method on our task. △ Less

Submitted 20 December, 2023; originally announced December 2023.

Comments: 8 pages, 4 figures

ACM Class: I.2.7; I.7.5

Journal ref: 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy, 2021, pp. 5543-5550

Can It Edit? Evaluating the Ability of Large Language Models to Follow Code Editing Instructions

Authors: Federico Cassano, Luisa Li, Akul Sethi, Noah Shinn, Abby Brennan-Jones, Jacob Ginesin, Edward Berman, George Chakhnashvili, Anton Lozhkov, Carolyn Jane Anderson, Arjun Guha

Abstract: A significant amount of research is focused on developing and evaluating large language models for a variety of code synthesis tasks. These include synthesizing code from natural language, synthesizing tests from code, and synthesizing explanations of code. In contrast, the behavior of instructional code editing with LLMs is understudied. These are tasks in which the model is provided a block of c… ▽ More A significant amount of research is focused on developing and evaluating large language models for a variety of code synthesis tasks. These include synthesizing code from natural language, synthesizing tests from code, and synthesizing explanations of code. In contrast, the behavior of instructional code editing with LLMs is understudied. These are tasks in which the model is provided a block of code and an instruction to modify the code. The editing instruction may ask for a feature to be added or removed, describe a bug and ask for a fix, or ask for a different kind of solution. We introduce a carefully crafted benchmark of code editing tasks and use it to evaluate several cutting edge LLMs. Our evaluation exposes a significant gap between the capabilities of state-of-the-art open and closed models. For example, even GPT-3.5-Turbo is better than the best open model at code editing tasks. We also introduce a new, carefully curated, permissively licensed training dataset of code editing tasks coupled with natural language instructions. Using this training dataset, we show that we can fine-tune open Code LLMs to significantly improve their code editing capabilities, closing the gap between open and closed models. All code, data, and models are available at https://github.com/nuprl/CanItEdit. △ Less

Submitted 23 September, 2024; v1 submitted 10 December, 2023; originally announced December 2023.

Exploring Emotional and Social Dynamics in Mobile Usage During Home Confinement

Authors: Nan Gao, Sam Nolan, Kaixin Ji, Shakila Khan Rumi, Judith Simone Heinisch, Christoph Anderson, Klaus David, Flora D. Salim

Abstract: Home confinement, a situation experienced by individuals for reasons ranging from medical quarantines, rehabilitation needs, disability accommodations, and remote working, is a common yet impactful aspect of modern life. While essential in various scenarios, confinement within the home environment can profoundly influence mental well-being and digital device usage. Using the COVID-19 lockdown as a… ▽ More Home confinement, a situation experienced by individuals for reasons ranging from medical quarantines, rehabilitation needs, disability accommodations, and remote working, is a common yet impactful aspect of modern life. While essential in various scenarios, confinement within the home environment can profoundly influence mental well-being and digital device usage. Using the COVID-19 lockdown as a case study, this research explores the emotional and social effects of prolonged home confinement on mobile device usage. We conducted an in-situ study with 32 participants, analyzing three weeks of mobile usage data to assess emotional well-being and social dynamics in restricted environments. Our findings reveal that app usage patterns serve as strong indicators of emotional states, offering insights into how digital interactions can reflect and influence well-being during isolation. This study highlights the potential for developing targeted interventions and support systems for individuals in long-term home confinement, including those with chronic illness, recovery needs, or permanent remote work situations. △ Less

Submitted 7 July, 2025; v1 submitted 20 October, 2023; originally announced October 2023.

Autonomous Systems' Safety Cases for use in UK Nuclear Environments

Authors: Christopher R. Anderson, Louise A. Dennis

Abstract: An overview of the process to develop a safety case for an autonomous robot deployment on a nuclear site in the UK is described and a safety case for a hypothetical robot incorporating AI is presented. This forms a first step towards a deployment, showing what is possible now and what may be possible with development of tools. It forms the basis for further discussion between nuclear site licensee… ▽ More An overview of the process to develop a safety case for an autonomous robot deployment on a nuclear site in the UK is described and a safety case for a hypothetical robot incorporating AI is presented. This forms a first step towards a deployment, showing what is possible now and what may be possible with development of tools. It forms the basis for further discussion between nuclear site licensees, the Office for Nuclear Regulation (ONR), industry and academia. △ Less

Submitted 3 October, 2023; originally announced October 2023.

Comments: In Proceedings AREA 2023, arXiv:2310.00333

Journal ref: EPTCS 391, 2023, pp. 83-88

A Large Language Model Approach to Educational Survey Feedback Analysis

Authors: Michael J. Parker, Caitlin Anderson, Claire Stone, YeaRim Oh

Abstract: This paper assesses the potential for the large language models (LLMs) GPT-4 and GPT-3.5 to aid in deriving insight from education feedback surveys. Exploration of LLM use cases in education has focused on teaching and learning, with less exploration of capabilities in education feedback analysis. Survey analysis in education involves goals such as finding gaps in curricula or evaluating teachers,… ▽ More This paper assesses the potential for the large language models (LLMs) GPT-4 and GPT-3.5 to aid in deriving insight from education feedback surveys. Exploration of LLM use cases in education has focused on teaching and learning, with less exploration of capabilities in education feedback analysis. Survey analysis in education involves goals such as finding gaps in curricula or evaluating teachers, often requiring time-consuming manual processing of textual responses. LLMs have the potential to provide a flexible means of achieving these goals without specialized machine learning models or fine-tuning. We demonstrate a versatile approach to such goals by treating them as sequences of natural language processing (NLP) tasks including classification (multi-label, multi-class, and binary), extraction, thematic analysis, and sentiment analysis, each performed by LLM. We apply these workflows to a real-world dataset of 2500 end-of-course survey comments from biomedical science courses, and evaluate a zero-shot approach (i.e., requiring no examples or labeled training data) across all tasks, reflecting education settings, where labeled data is often scarce. By applying effective prompting practices, we achieve human-level performance on multiple tasks with GPT-4, enabling workflows necessary to achieve typical goals. We also show the potential of inspecting LLMs' chain-of-thought (CoT) reasoning for providing insight that may foster confidence in practice. Moreover, this study features development of a versatile set of classification categories, suitable for various course types (online, hybrid, or in-person) and amenable to customization. Our results suggest that LLMs can be used to derive a range of insights from survey text. △ Less

Submitted 26 June, 2024; v1 submitted 29 September, 2023; originally announced September 2023.

Journal ref: Int J Artif Intell Educ (2024)

Julia as a unifying end-to-end workflow language on the Frontier exascale system

Authors: William F. Godoy, Pedro Valero-Lara, Caira Anderson, Katrina W. Lee, Ana Gainaru, Rafael Ferreira da Silva, Jeffrey S. Vetter

Abstract: We evaluate Julia as a single language and ecosystem paradigm powered by LLVM to develop workflow components for high-performance computing. We run a Gray-Scott, 2-variable diffusion-reaction application using a memory-bound, 7-point stencil kernel on Frontier, the US Department of Energy's first exascale supercomputer. We evaluate the performance, scaling, and trade-offs of (i) the computational… ▽ More We evaluate Julia as a single language and ecosystem paradigm powered by LLVM to develop workflow components for high-performance computing. We run a Gray-Scott, 2-variable diffusion-reaction application using a memory-bound, 7-point stencil kernel on Frontier, the US Department of Energy's first exascale supercomputer. We evaluate the performance, scaling, and trade-offs of (i) the computational kernel on AMD's MI250x GPUs, (ii) weak scaling up to 4,096 MPI processes/GPUs or 512 nodes, (iii) parallel I/O writes using the ADIOS2 library bindings, and (iv) Jupyter Notebooks for interactive analysis. Results suggest that although Julia generates a reasonable LLVM-IR, a nearly 50% performance difference exists vs. native AMD HIP stencil codes when running on the GPUs. As expected, we observed near-zero overhead when using MPI and parallel I/O bindings for system-wide installed implementations. Consequently, Julia emerges as a compelling high-performance and high-productivity workflow composition language, as measured on the fastest supercomputer in the world. △ Less

Submitted 27 September, 2023; v1 submitted 18 September, 2023; originally announced September 2023.

Comments: 11 pages, 8 figures, accepted at the 18th Workshop on Workflows in Support of Large-Scale Science (WORKS23), IEEE/ACM The International Conference for High Performance Computing, Networking, Storage, and Analysis, SC23

Knowledge Transfer from High-Resource to Low-Resource Programming Languages for Code LLMs

Authors: Federico Cassano, John Gouwar, Francesca Lucchetti, Claire Schlesinger, Anders Freeman, Carolyn Jane Anderson, Molly Q Feldman, Michael Greenberg, Abhinav Jangda, Arjun Guha

Abstract: Over the past few years, Large Language Models of Code (Code LLMs) have started to have a significant impact on programming practice. Code LLMs are also emerging as building blocks for research in programming languages and software engineering. However, Code LLMs produce impressive results on programming languages that are well represented in their training data (e.g., Java, Python, or JavaScript)… ▽ More Over the past few years, Large Language Models of Code (Code LLMs) have started to have a significant impact on programming practice. Code LLMs are also emerging as building blocks for research in programming languages and software engineering. However, Code LLMs produce impressive results on programming languages that are well represented in their training data (e.g., Java, Python, or JavaScript), but struggle with low-resource languages that have limited training data available. Low resource languages include OCaml, Racket, and several others. This paper presents an effective approach for boosting the performance of Code LLMs on low-resource languages using semi-synthetic data. Our approach, MultiPL-T, translates training data from high-resource languages into training data for low-resource languages in the following way. 1) We use a Code LLM to synthesize tests for commented code from a high-resource language, filtering out faulty tests and code with low test coverage. 2) We use a Code LLM to translate Python code to a target low-resource language, and use tests to validate the translation. We apply this approach to generate tens of thousands of validated training items for Julia, Lua, OCaml, R, and Racket. Furthermore, we use an open model (StarCoderBase) with open training data (The Stack), which allows us to decontaminate benchmarks, train models without violating licenses, and run experiments that could not otherwise be done. With MultiPL-T generated data, we present fine-tuned versions of StarCoderBase and Code Llama for Julia, Lua, OCaml, R, and Racket. On established benchmarks (MultiPL-E), these models outperform other open Code LLMs. The MultiPL-T approach is easy to apply to new languages, and is significantly more efficient and effective than alternatives such as training longer. △ Less

Submitted 21 September, 2024; v1 submitted 18 August, 2023; originally announced August 2023.

A Multiobjective Reinforcement Learning Framework for Microgrid Energy Management

Authors: M. Vivienne Liu, Patrick M. Reed, David Gold, Garret Quist, C. Lindsay Anderson

Abstract: The emergence of microgrids (MGs) has provided a promising solution for decarbonizing and decentralizing the power grid, mitigating the challenges posed by climate change. However, MG operations often involve considering multiple objectives that represent the interests of different stakeholders, leading to potentially complex conflicts. To tackle this issue, we propose a novel multi-objective rein… ▽ More The emergence of microgrids (MGs) has provided a promising solution for decarbonizing and decentralizing the power grid, mitigating the challenges posed by climate change. However, MG operations often involve considering multiple objectives that represent the interests of different stakeholders, leading to potentially complex conflicts. To tackle this issue, we propose a novel multi-objective reinforcement learning framework that explores the high-dimensional objective space and uncovers the tradeoffs between conflicting objectives. This framework leverages exogenous information and capitalizes on the data-driven nature of reinforcement learning, enabling the training of a parametric policy without the need for long-term forecasts or knowledge of the underlying uncertainty distribution. The trained policies exhibit diverse, adaptive, and coordinative behaviors with the added benefit of providing interpretable insights on the dynamics of their information use. We employ this framework on the Cornell University MG (CU-MG), which is a combined heat and power MG, to evaluate its effectiveness. The results demonstrate performance improvements in all objectives considered compared to the status quo operations and offer more flexibility in navigating complex operational tradeoffs. △ Less

Submitted 14 February, 2025; v1 submitted 17 July, 2023; originally announced July 2023.

Solving and Generating NPR Sunday Puzzles with Large Language Models

Authors: Jingmiao Zhao, Carolyn Jane Anderson

Abstract: We explore the ability of large language models to solve and generate puzzles from the NPR Sunday Puzzle game show using PUZZLEQA, a dataset comprising 15 years of on-air puzzles. We evaluate four large language models using PUZZLEQA, in both multiple choice and free response formats, and explore two prompt engineering techniques to improve free response performance: chain-of-thought reasoning and… ▽ More We explore the ability of large language models to solve and generate puzzles from the NPR Sunday Puzzle game show using PUZZLEQA, a dataset comprising 15 years of on-air puzzles. We evaluate four large language models using PUZZLEQA, in both multiple choice and free response formats, and explore two prompt engineering techniques to improve free response performance: chain-of-thought reasoning and prompt summarization. We find that state-of-the-art large language models can solve many PUZZLEQA puzzles: the best model, GPT-3.5, achieves 50.2% loose accuracy. However, in our few-shot puzzle generation experiment, we find no evidence that models can generate puzzles: GPT-3.5 generates puzzles with answers that do not conform to the generated rules. Puzzle generation remains a challenging task for future work. △ Less

Submitted 21 June, 2023; originally announced June 2023.

Comments: To appear in the Proceedings of the 14th International Conference on Computational Creativity (ICCC)

StudentEval: A Benchmark of Student-Written Prompts for Large Language Models of Code

Authors: Hannah McLean Babe, Sydney Nguyen, Yangtian Zi, Arjun Guha, Molly Q Feldman, Carolyn Jane Anderson

Abstract: Code LLMs are being rapidly deployed and there is evidence that they can make professional programmers more productive. Current benchmarks for code generation measure whether models generate correct programs given an expert prompt. In this paper, we present a new benchmark containing multiple prompts per problem, written by a specific population of non-expert prompters: beginning programmers. Stud… ▽ More Code LLMs are being rapidly deployed and there is evidence that they can make professional programmers more productive. Current benchmarks for code generation measure whether models generate correct programs given an expert prompt. In this paper, we present a new benchmark containing multiple prompts per problem, written by a specific population of non-expert prompters: beginning programmers. StudentEval contains 1,749 prompts for 48 problems, written by 80 students who have only completed one semester of Python programming. Our students wrote these prompts while working interactively with a Code LLM, and we observed very mixed success rates. We use StudentEval to evaluate 5 Code LLMs and find that StudentEval is a better discriminator of model performance than existing benchmarks. We analyze the prompts and find significant variation in students' prompting techniques. We also find that nondeterministic LLM sampling could mislead students into thinking that their prompts are more (or less) effective than they actually are, which has implications for how to teach with Code LLMs. △ Less

Submitted 7 June, 2023; originally announced June 2023.

Keep It Simple: Fault Tolerance Evaluation of Federated Learning with Unreliable Clients

Authors: Victoria Huang, Shaleeza Sohail, Michael Mayo, Tania Lorido Botran, Mark Rodrigues, Chris Anderson, Melanie Ooi

Abstract: Federated learning (FL), as an emerging artificial intelligence (AI) approach, enables decentralized model training across multiple devices without exposing their local training data. FL has been increasingly gaining popularity in both academia and industry. While research works have been proposed to improve the fault tolerance of FL, the real impact of unreliable devices (e.g., dropping out, misc… ▽ More Federated learning (FL), as an emerging artificial intelligence (AI) approach, enables decentralized model training across multiple devices without exposing their local training data. FL has been increasingly gaining popularity in both academia and industry. While research works have been proposed to improve the fault tolerance of FL, the real impact of unreliable devices (e.g., dropping out, misconfiguration, poor data quality) in real-world applications is not fully investigated. We carefully chose two representative, real-world classification problems with a limited numbers of clients to better analyze FL fault tolerance. Contrary to the intuition, simple FL algorithms can perform surprisingly well in the presence of unreliable clients. △ Less

Submitted 16 May, 2023; originally announced May 2023.

StarCoder: may the source be with you!

Authors: Raymond Li, Loubna Ben Allal, Yangtian Zi, Niklas Muennighoff, Denis Kocetkov, Chenghao Mou, Marc Marone, Christopher Akiki, Jia Li, Jenny Chim, Qian Liu, Evgenii Zheltonozhskii, Terry Yue Zhuo, Thomas Wang, Olivier Dehaene, Mishig Davaadorj, Joel Lamy-Poirier, João Monteiro, Oleh Shliazhko, Nicolas Gontier, Nicholas Meade, Armel Zebaze, Ming-Ho Yee, Logesh Kumar Umapathi, Jian Zhu , et al. (42 additional authors not shown)

Abstract: The BigCode community, an open-scientific collaboration working on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder and StarCoderBase: 15.5B parameter models with 8K context length, infilling capabilities and fast large-batch inference enabled by multi-query attention. StarCoderBase is trained on 1 trillion tokens sourced from The Stack, a large colle… ▽ More The BigCode community, an open-scientific collaboration working on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder and StarCoderBase: 15.5B parameter models with 8K context length, infilling capabilities and fast large-batch inference enabled by multi-query attention. StarCoderBase is trained on 1 trillion tokens sourced from The Stack, a large collection of permissively licensed GitHub repositories with inspection tools and an opt-out process. We fine-tuned StarCoderBase on 35B Python tokens, resulting in the creation of StarCoder. We perform the most comprehensive evaluation of Code LLMs to date and show that StarCoderBase outperforms every open Code LLM that supports multiple programming languages and matches or outperforms the OpenAI code-cushman-001 model. Furthermore, StarCoder outperforms every model that is fine-tuned on Python, can be prompted to achieve 40\% pass@1 on HumanEval, and still retains its performance on other programming languages. We take several important steps towards a safe open-access model release, including an improved PII redaction pipeline and a novel attribution tracing tool, and make the StarCoder models publicly available under a more commercially viable version of the Open Responsible AI Model license. △ Less

Submitted 13 December, 2023; v1 submitted 9 May, 2023; originally announced May 2023.

Propagation Measurements and Analyses at 28 GHz via an Autonomous Beam-Steering Platform

Authors: Bharath Keshavamurthy, Yaguang Zhang, Christopher R. Anderson, Nicolo Michelusi, James V. Krogmeier, David J. Love

Abstract: This paper details the design of an autonomous alignment and tracking platform to mechanically steer directional horn antennas in a sliding correlator channel sounder setup for 28 GHz V2X propagation modeling. A pan-and-tilt subsystem facilitates uninhibited rotational mobility along the yaw and pitch axes, driven by open-loop servo units and orchestrated via inertial motion controllers. A geo-pos… ▽ More This paper details the design of an autonomous alignment and tracking platform to mechanically steer directional horn antennas in a sliding correlator channel sounder setup for 28 GHz V2X propagation modeling. A pan-and-tilt subsystem facilitates uninhibited rotational mobility along the yaw and pitch axes, driven by open-loop servo units and orchestrated via inertial motion controllers. A geo-positioning subsystem augmented in accuracy by real-time kinematics enables navigation events to be shared between a transmitter and receiver over an Apache Kafka messaging middleware framework with fault tolerance. Herein, our system demonstrates a 3D geo-positioning accuracy of 17 cm, an average principal axes positioning accuracy of 1.1 degrees, and an average tracking response time of 27.8 ms. Crucially, fully autonomous antenna alignment and tracking facilitates continuous series of measurements, a unique yet critical necessity for millimeter wave channel modeling in vehicular networks. The power-delay profiles, collected along routes spanning urban and suburban neighborhoods on the NSF POWDER testbed, are used in pathloss evaluations involving the 3GPP TR38.901 and ITU-R M.2135 standards. Empirically, we demonstrate that these models fail to accurately capture the 28 GHz pathloss behavior in urban foliage and suburban radio environments. In addition to RMS direction-spread analyses for angles-of-arrival via the SAGE algorithm, we perform signal decoherence studies wherein we derive exponential models for the spatial/angular autocorrelation coefficient under distance and alignment effects. △ Less

Submitted 16 February, 2023; originally announced February 2023.

Comments: 6 pages, 18 figures, 2 tables; Accepted at IEEE International Conference on Communications (ICC) 2023: Paper #1570867736

Report number: ICC Paper #1570867736

SantaCoder: don't reach for the stars!

Authors: Loubna Ben Allal, Raymond Li, Denis Kocetkov, Chenghao Mou, Christopher Akiki, Carlos Munoz Ferrandis, Niklas Muennighoff, Mayank Mishra, Alex Gu, Manan Dey, Logesh Kumar Umapathi, Carolyn Jane Anderson, Yangtian Zi, Joel Lamy Poirier, Hailey Schoelkopf, Sergey Troshin, Dmitry Abulkhanov, Manuel Romero, Michael Lappert, Francesco De Toni, Bernardo García del Río, Qian Liu, Shamik Bose, Urvashi Bhattacharyya, Terry Yue Zhuo , et al. (16 additional authors not shown)

Abstract: The BigCode project is an open-scientific collaboration working on the responsible development of large language models for code. This tech report describes the progress of the collaboration until December 2022, outlining the current state of the Personally Identifiable Information (PII) redaction pipeline, the experiments conducted to de-risk the model architecture, and the experiments investigat… ▽ More The BigCode project is an open-scientific collaboration working on the responsible development of large language models for code. This tech report describes the progress of the collaboration until December 2022, outlining the current state of the Personally Identifiable Information (PII) redaction pipeline, the experiments conducted to de-risk the model architecture, and the experiments investigating better preprocessing methods for the training data. We train 1.1B parameter models on the Java, JavaScript, and Python subsets of The Stack and evaluate them on the MultiPL-E text-to-code benchmark. We find that more aggressive filtering of near-duplicates can further boost performance and, surprisingly, that selecting files from repositories with 5+ GitHub stars deteriorates performance significantly. Our best model outperforms previous open-source multilingual code generation models (InCoder-6.7B and CodeGen-Multi-2.7B) in both left-to-right generation and infilling on the Java, JavaScript, and Python portions of MultiPL-E, despite being a substantially smaller model. All models are released under an OpenRAIL license at https://hf.co/bigcode. △ Less

Submitted 24 February, 2023; v1 submitted 9 January, 2023; originally announced January 2023.

An Interpretable Model of Climate Change Using Correlative Learning

Authors: Charles Anderson, Jason Stock

Abstract: Determining changes in global temperature and precipitation that may indicate climate change is complicated by annual variations. One approach for finding potential climate change indicators is to train a model that predicts the year from annual means of global temperatures and precipitations. Such data is available from the CMIP6 ensemble of simulations. Here a two-hidden-layer neural network tra… ▽ More Determining changes in global temperature and precipitation that may indicate climate change is complicated by annual variations. One approach for finding potential climate change indicators is to train a model that predicts the year from annual means of global temperatures and precipitations. Such data is available from the CMIP6 ensemble of simulations. Here a two-hidden-layer neural network trained on this data successfully predicts the year. Differences among temperature and precipitation patterns for which the model predicts specific years reveal changes through time. To find these optimal patterns, a new way of interpreting what the neural network has learned is explored. Alopex, a stochastic correlative learning algorithm, is used to find optimal temperature and precipitation maps that best predict a given year. These maps are compared over multiple years to show how temperature and precipitations patterns indicative of each year change over time. △ Less

Submitted 5 December, 2022; originally announced December 2022.

Comments: NeurIPS 2022 Workshop - Tackling Climate Change with Machine Learning, 4 page limit w/ appendix

Attention-Based Scattering Network for Satellite Imagery

Authors: Jason Stock, Chuck Anderson

Abstract: Multi-channel satellite imagery, from stacked spectral bands or spatiotemporal data, have meaningful representations for various atmospheric properties. Combining these features in an effective manner to create a performant and trustworthy model is of utmost importance to forecasters. Neural networks show promise, yet suffer from unintuitive computations, fusion of high-level features, and may be… ▽ More Multi-channel satellite imagery, from stacked spectral bands or spatiotemporal data, have meaningful representations for various atmospheric properties. Combining these features in an effective manner to create a performant and trustworthy model is of utmost importance to forecasters. Neural networks show promise, yet suffer from unintuitive computations, fusion of high-level features, and may be limited by the quantity of available data. In this work, we leverage the scattering transform to extract high-level features without additional trainable parameters and introduce a separation scheme to bring attention to independent input channels. Experiments show promising results on estimating tropical cyclone intensity and predicting the occurrence of lightning from satellite imagery. △ Less

Submitted 21 October, 2022; originally announced October 2022.

Comments: NeurIPS 2022 Workshop - Tackling Climate Change with Machine Learning, 4 page limit w/ appendix

MultiPL-E: A Scalable and Extensible Approach to Benchmarking Neural Code Generation

Authors: Federico Cassano, John Gouwar, Daniel Nguyen, Sydney Nguyen, Luna Phipps-Costin, Donald Pinckney, Ming-Ho Yee, Yangtian Zi, Carolyn Jane Anderson, Molly Q Feldman, Arjun Guha, Michael Greenberg, Abhinav Jangda

Abstract: Large language models have demonstrated the ability to generate both natural language and programming language text. Such models open up the possibility of multi-language code generation: could code generation models generalize knowledge from one language to another? Although contemporary code generation models can generate semantically correct Python code, little is known about their abilities wi… ▽ More Large language models have demonstrated the ability to generate both natural language and programming language text. Such models open up the possibility of multi-language code generation: could code generation models generalize knowledge from one language to another? Although contemporary code generation models can generate semantically correct Python code, little is known about their abilities with other languages. We propose MultiPL-E, a system for translating unit test-driven code generation benchmarks to new languages. We create the first massively multilingual code generation benchmark by using MultiPL-E to translate two popular Python code generation benchmarks to 18 additional programming languages. We use MultiPL-E to extend the HumanEval benchmark and MBPP benchmark to 18 languages that encompass a range of programming paradigms and popularity. Using these new parallel benchmarks, we evaluate the multi-language performance of three state-of-the-art code generation models: Codex, CodeGen, and InCoder. We find that Codex matches or even exceeds its performance on Python for several other languages. The range of programming languages represented in MultiPL-E allow us to explore the impact of language frequency and language features on model performance. Finally, the MultiPL-E approach of compiling code generation benchmarks to new programming languages is both scalable and extensible, making it straightforward to evaluate new models, benchmarks, and languages. △ Less

Submitted 19 December, 2022; v1 submitted 17 August, 2022; originally announced August 2022.

Investigating the Effects of Mood & Usage Behaviour on Notification Response Time

Authors: Judith S. Heinisch, Nan Gao, Christoph Anderson, Shohreh Deldari, Klaus David, Flora Salim

Abstract: Notifications are one of the most prevailing mechanisms on smartphones and personal computers to convey timely and important information. Despite these benefits, smartphone notifications demand individuals' attention and can cause stress and frustration when delivered at inopportune timings. This paper investigates the effect of individuals' smartphone usage behavior and mood on notification respo… ▽ More Notifications are one of the most prevailing mechanisms on smartphones and personal computers to convey timely and important information. Despite these benefits, smartphone notifications demand individuals' attention and can cause stress and frustration when delivered at inopportune timings. This paper investigates the effect of individuals' smartphone usage behavior and mood on notification response time. We conduct an in-the-wild study with more than 18 participants for five weeks. Extensive experiment results show that the proposed regression model is able to accurately predict the response time of smartphone notifications using current user's mood and physiological signals. We explored the effect of different features for each participant to choose the most important user-oriented features in order to to achieve a meaningful and personalised notification response prediction. On average, our regression model achieved over all participants an MAE of 0.7764 ms and RMSE of 1.0527 ms. We also investigate how physiological signals (collected from E4 wristbands) are used as an indicator for mood and discuss the individual differences in application usage and categories of smartphone applications on the response time of notifications. Our research sheds light on the future intelligent notification management system. △ Less

Submitted 7 July, 2022; originally announced July 2022.

Interpretable Climate Change Modeling With Progressive Cascade Networks

Authors: Charles Anderson, Jason Stock, David Anderson

Abstract: Typical deep learning approaches to modeling high-dimensional data often result in complex models that do not easily reveal a new understanding of the data. Research in the deep learning field is very actively pursuing new methods to interpret deep neural networks and to reduce their complexity. An approach is described here that starts with linear models and incrementally adds complexity only as… ▽ More Typical deep learning approaches to modeling high-dimensional data often result in complex models that do not easily reveal a new understanding of the data. Research in the deep learning field is very actively pursuing new methods to interpret deep neural networks and to reduce their complexity. An approach is described here that starts with linear models and incrementally adds complexity only as supported by the data. An application is shown in which models that map global temperature and precipitation to years are trained to investigate patterns associated with changes in climate. △ Less

Submitted 12 May, 2022; originally announced May 2022.

Trainable Wavelet Neural Network for Non-Stationary Signals

Authors: Jason Stock, Chuck Anderson

Abstract: This work introduces a wavelet neural network to learn a filter-bank specialized to fit non-stationary signals and improve interpretability and performance for digital signal processing. The network uses a wavelet transform as the first layer of a neural network where the convolution is a parameterized function of the complex Morlet wavelet. Experimental results, on both simplified data and atmosp… ▽ More This work introduces a wavelet neural network to learn a filter-bank specialized to fit non-stationary signals and improve interpretability and performance for digital signal processing. The network uses a wavelet transform as the first layer of a neural network where the convolution is a parameterized function of the complex Morlet wavelet. Experimental results, on both simplified data and atmospheric gravity waves, show the network is quick to converge, generalizes well on noisy data, and outperforms standard network architectures. △ Less

Submitted 6 May, 2022; originally announced May 2022.

Comments: AI for Earth and Space Science Workshop at the International Conference on Learning Representations (ICLR), April, 2022

Initial Investigations Towards Non-invasive Monitoring of Chronic Wound Healing Using Deep Learning and Ultrasound Imaging

Authors: Maja Schlereth, Daniel Stromer, Yash Mantri, Jason Tsujimoto, Katharina Breininger, Andreas Maier, Caesar Anderson, Pranav S. Garimella, Jesse V. Jokerst

Abstract: Chronic wounds including diabetic and arterial/venous insufficiency injuries have become a major burden for healthcare systems worldwide. Demographic changes suggest that wound care will play an even bigger role in the coming decades. Predicting and monitoring response to therapy in wound care is currently largely based on visual inspection with little information on the underlying tissue. Thus, t… ▽ More Chronic wounds including diabetic and arterial/venous insufficiency injuries have become a major burden for healthcare systems worldwide. Demographic changes suggest that wound care will play an even bigger role in the coming decades. Predicting and monitoring response to therapy in wound care is currently largely based on visual inspection with little information on the underlying tissue. Thus, there is an urgent unmet need for innovative approaches that facilitate personalized diagnostics and treatments at the point-of-care. It has been recently shown that ultrasound imaging can monitor response to therapy in wound care, but this work required onerous manual image annotations. In this study, we present initial results of a deep learning-based automatic segmentation of cross-sectional wound size in ultrasound images and identify requirements and challenges for future research on this application. Evaluation of the segmentation results underscores the potential of the proposed deep learning approach to complement non-invasive imaging with Dice scores of 0.34 (U-Net, FCN) and 0.27 (ResNet-U-Net) but also highlights the need for improving robustness further. We conclude that deep learning-supported analysis of non-invasive ultrasound images is a promising area of research to automatically extract cross-sectional wound size and depth information with potential value in monitoring response to therapy. △ Less

Submitted 25 January, 2022; originally announced January 2022.

Comments: 6 pages, 2 figures, accepted by BVM conference proceedings 2022

Statistical detection of format dialects using the weighted Dowker complex

Authors: Michael Robinson, Letitia W. Li, Cory Anderson, Steve Huntsman

Abstract: This paper provides an experimentally validated, probabilistic model of file behavior when consumed by a set of pre-existing parsers. File behavior is measured by way of a standardized set of Boolean "messages" produced as the files are read. By thresholding the posterior probability that a file exhibiting a particular set of messages is from a particular dialect, our model yields a practical clas… ▽ More This paper provides an experimentally validated, probabilistic model of file behavior when consumed by a set of pre-existing parsers. File behavior is measured by way of a standardized set of Boolean "messages" produced as the files are read. By thresholding the posterior probability that a file exhibiting a particular set of messages is from a particular dialect, our model yields a practical classification algorithm for two dialects. We demonstrate that this thresholding algorithm for two dialects can be bootstrapped from a training set consisting primarily of one dialect. Both the (parametric) theoretical and the (non-parametric) empirical distributions of file behaviors for one dialect yield good classification performance, and outperform classification based on simply counting messages. Our theoretical framework relies on statistical independence of messages within each dialect. Violations of this assumption are detectable and allow a format analyst to identify "boundaries" between dialects. A format analyst can therefore greatly reduce the number of files they need to consider when crafting new criteria for dialect detection, since they need only consider the files that exhibit ambiguous message patterns. △ Less

Submitted 20 January, 2022; originally announced January 2022.

Comments: 15 pages, 11 figures, 5 tables

MSC Class: 62P30; 55U10 ACM Class: D.3.4

Automatic Extraction of Medication Names in Tweets as Named Entity Recognition

Authors: Carol Anderson, Bo Liu, Anas Abidin, Hoo-Chang Shin, Virginia Adams

Abstract: Social media posts contain potentially valuable information about medical conditions and health-related behavior. Biocreative VII Task 3 focuses on mining this information by recognizing mentions of medications and dietary supplements in tweets. We approach this task by fine tuning multiple BERT-style language models to perform token-level classification, and combining them into ensembles to gener… ▽ More Social media posts contain potentially valuable information about medical conditions and health-related behavior. Biocreative VII Task 3 focuses on mining this information by recognizing mentions of medications and dietary supplements in tweets. We approach this task by fine tuning multiple BERT-style language models to perform token-level classification, and combining them into ensembles to generate final predictions. Our best system consists of five Megatron-BERT-345M models and achieves a strict F1 score of 0.764 on unseen test data. △ Less

Submitted 30 November, 2021; originally announced November 2021.

Comments: Submission to the BioCreative VII challenge - Track-3

Chemical Identification and Indexing in PubMed Articles via BERT and Text-to-Text Approaches

Authors: Virginia Adams, Hoo-Chang Shin, Carol Anderson, Bo Liu, Anas Abidin

Abstract: The Biocreative VII Track-2 challenge consists of named entity recognition, entity-linking (or entity-normalization), and topic indexing tasks -- with entities and topics limited to chemicals for this challenge. Named entity recognition is a well-established problem and we achieve our best performance with BERT-based BioMegatron models. We extend our BERT-based approach to the entity linking task.… ▽ More The Biocreative VII Track-2 challenge consists of named entity recognition, entity-linking (or entity-normalization), and topic indexing tasks -- with entities and topics limited to chemicals for this challenge. Named entity recognition is a well-established problem and we achieve our best performance with BERT-based BioMegatron models. We extend our BERT-based approach to the entity linking task. After the second stage of pretraining BioBERT with a metric-learning loss strategy called self-alignment pretraining (SAP), we link entities based on the cosine similarity between their SAP-BioBERT word embeddings. Despite the success of our named entity recognition experiments, we find the chemical indexing task generally more challenging. In addition to conventional NER methods, we attempt both named entity recognition and entity linking with a novel text-to-text or "prompt" based method that uses generative language models such as T5 and GPT. We achieve encouraging results with this new approach. △ Less

Submitted 30 November, 2021; originally announced November 2021.

Comments: Submission to the BioCreative VII challenge - Track-2

Text Mining Drug/Chemical-Protein Interactions using an Ensemble of BERT and T5 Based Models

Authors: Virginia Adams, Hoo-Chang Shin, Carol Anderson, Bo Liu, Anas Abidin

Abstract: In Track-1 of the BioCreative VII Challenge participants are asked to identify interactions between drugs/chemicals and proteins. In-context named entity annotations for each drug/chemical and protein are provided and one of fourteen different interactions must be automatically predicted. For this relation extraction task, we attempt both a BERT-based sentence classification approach, and a more n… ▽ More In Track-1 of the BioCreative VII Challenge participants are asked to identify interactions between drugs/chemicals and proteins. In-context named entity annotations for each drug/chemical and protein are provided and one of fourteen different interactions must be automatically predicted. For this relation extraction task, we attempt both a BERT-based sentence classification approach, and a more novel text-to-text approach using a T5 model. We find that larger BERT-based models perform better in general, with our BioMegatron-based model achieving the highest scores across all metrics, achieving 0.74 F1 score. Though our novel T5 text-to-text method did not perform as well as most of our BERT-based models, it outperformed those trained on similar data, showing promising results, achieving 0.65 F1 score. We believe a text-to-text approach to relation extraction has some competitive advantages and there is a lot of room for research advancement. △ Less

Submitted 30 November, 2021; originally announced November 2021.

Comments: Submission to the BioCreative VII challenge, Track-1