Showing 1–50 of 180 results for author: Clark, J

Automated Stealthy Wear-Out Attack on Digital Twins With Deep Reinforcement Learning

Authors: Joshua Haworth, Aryan Pasikhani, George Pavlides, Prosanta Gope, John Clark

Abstract: Digital Twins (DTs) have emerged as pivotal enablers of Industry 4.0, offering transformative capabilities such as real-time monitoring, advanced simulation, and precise control of physical assets. By bridging the physical and virtual domains, DTs facilitate seamless integration of data-driven decision-making and operational optimisation. However, this seamless interaction significantly expands th… ▽ More Digital Twins (DTs) have emerged as pivotal enablers of Industry 4.0, offering transformative capabilities such as real-time monitoring, advanced simulation, and precise control of physical assets. By bridging the physical and virtual domains, DTs facilitate seamless integration of data-driven decision-making and operational optimisation. However, this seamless interaction significantly expands the attack surface of industrial systems, creating vulnerabilities that adversaries can exploit. This paper introduces a novel and stealthy wear-out attack leveraging Deep Reinforcement Learning (DRL) to target DT-enabled infrastructures. The adversary strategically and covertly manipulates control signals, inducing increased torque on a specific joint to accelerate wear and tear while evading detection by a state-of-the-art anomaly detection system. Extensive benchmarking of reinforcement learning algorithms - including Twin Delayed Deep Deterministic Policy Gradient (TD3), Soft Actor-Critic (SAC), Proximal Policy Optimisation (PPO), and Advantage Actor-Critic (A2C) - revealed that SAC consistently outperformed its counterparts in terms of sample efficiency, stability, and overall attack effectiveness. We evaluate the proposed adversary in an industrial setting using the UR10e robotic arm. Results demonstrate the adversary's ability to significantly elevate torque levels on the targeted joint, leading to accelerated degradation and increased maintenance costs, all while operating stealthily and avoiding detection. Our findings highlight the substantial risks posed by DRL-driven adversaries to DT-enabled environments and emphasise the critical need for robust defence mechanisms to protect critical industrial systems. △ Less

Submitted 12 July, 2026; originally announced July 2026.

Evaluation and Explainability of Unsupervised Scholarly Collaboration Recommendations

Authors: Md Asaduzzaman Noor, John W. Sheppard, Jason A. Clark

Abstract: In this paper, we examine unsupervised, content-based collaboration recommendations using publication text in scholarly settings. We compare three families of methods: a TF-IDF baseline, topic-based models (LDA and BERTopic, including clone variants), and embedding-based retrieval using SciBERT with Faiss. To evaluate model behavior beyond simple lexical matching, we introduce a constrained settin… ▽ More In this paper, we examine unsupervised, content-based collaboration recommendations using publication text in scholarly settings. We compare three families of methods: a TF-IDF baseline, topic-based models (LDA and BERTopic, including clone variants), and embedding-based retrieval using SciBERT with Faiss. To evaluate model behavior beyond simple lexical matching, we introduce a constrained setting where publication overlap between researchers is partially removed while still using historical co-authorship as proxy ground truth for post-hoc evaluation. Results show clear differences across methods. TF-IDF performs best under full information but drops significantly as overlap is reduced. In contrast, topic-based and embedding-based approaches show more stable performance, suggesting they capture broader distributional similarities, rather than relying only on direct lexical overlap. We also examine explainability through two perspectives: intrinsic topic-based explanations and post-hoc, retrieval-based explanations generated using language models. These provide complementary trade-offs between transparency and human readability. △ Less

Submitted 2 August, 2026; v1 submitted 5 July, 2026; originally announced July 2026.

Comments: 6 pages, 2 figures, Submitted to ICMLA 2026

Multisensory Continual Learning: Adapting Pretrained Visuomotor Policies to Force

Authors: Jaden Clark, Changhao Wang, Yihuai Gao, Seongheon Hong, Hojung Choi, Mark Cutkosky, Yifan Hou, Shuran Song

Abstract: Robot manipulation often relies on sensory feedback beyond vision, particularly in contact-rich settings where force, tactile, or audio signals reveal interaction states that are not directly observable from images. However, these modalities are often hardware- and task-specific, and large-scale multisensory robot datasets remain scarce. As a result, it is impractical to pretrain policies with eve… ▽ More Robot manipulation often relies on sensory feedback beyond vision, particularly in contact-rich settings where force, tactile, or audio signals reveal interaction states that are not directly observable from images. However, these modalities are often hardware- and task-specific, and large-scale multisensory robot datasets remain scarce. As a result, it is impractical to pretrain policies with every sensor they may encounter. We study multisensory continual learning: adapting a pretrained robot policy to new tasks with newly introduced modalities while preserving performance under the original sensor suite. We propose MultiSensory World Model (MuSe), which incorporates limited multisensory data into pretrained vision-only policies through multi-stage fusion, multisensory future prediction, and experience replay over pretraining data. We instantiate MuSe by augmenting a pretrained vision-only policy with force-torque sensing and evaluate it on real-world manipulation tasks. Our experiments show that MuSe performs strongly on contact-rich finetuning tasks while preserving, and in some cases improving, performance on the original pretraining tasks. These results suggest that a modest multisensory dataset can improve general robot capabilities beyond the finetuning distribution. Project website: https://jadenvc.github.io/multisensory-continual-learning/ △ Less

Submitted 5 July, 2026; v1 submitted 29 June, 2026; originally announced June 2026.

Artificial Intelligence Index Report 2026

Authors: Sha Sajadieh, Loredana Fattorini, Raymond Perrault, Yolanda Gil, Vanessa Parli, Lapo Santarlasci, Juan Pava, Nestor Maslej, Russ Altman, Erik Brynjolfsson, Carla Brodley, Jack Clark, Virginia Dignum, Vipin Kumar, James Landay, Terah Lyons, James Manyika, Juan Carlos Niebles, Yoav Shoham, Elham Tabassi, Russell Wald, Toby Walsh, Dan Weld

Abstract: Welcome to the ninth edition of the AI Index report. As AI continues to advance rapidly, the question becomes whether the systems built around it can keep up. Governance frameworks, evaluation methods, education systems, and the data infrastructure needed to track AI's impact are struggling to match the pace of the technology itself. That gap between what AI can do and how prepared we are to manag… ▽ More Welcome to the ninth edition of the AI Index report. As AI continues to advance rapidly, the question becomes whether the systems built around it can keep up. Governance frameworks, evaluation methods, education systems, and the data infrastructure needed to track AI's impact are struggling to match the pace of the technology itself. That gap between what AI can do and how prepared we are to manage it runs through every chapter of this year's report. New in this edition, the report tracks how AI is being tested more ambitiously across reasoning, safety, and real-world task execution, and why those measurements are increasingly difficult to rely on. It also features new estimates of generative AI's economic value alongside emerging evidence of its labor market effects, an analytical framework on AI sovereignty, and a science chapter developed in collaboration with Schmidt Sciences. For the first time, the report features standalone chapters on AI in science and AI in medicine, reflecting AI's growing impact across these two domains. △ Less

Submitted 29 June, 2026; v1 submitted 13 April, 2026; originally announced June 2026.

Model-based Optimization of Anguilliform Swimming Gaits for Soft Robotic Applications

Authors: Brian Van Stratum, James Gallentine, Caleb Rucker, Eric Barth, Jonathan E. Clark, Kourosh Shoele

Abstract: In this paper, we introduce the Soft Lamprey-Inspired Dual Environment Robot (SLIDER) and a proper modeling and optimization procedure employed to design the robot. We represent the primary fluid environment actions - inertial effects, vortex forces, and viscous dissipation - using Lighthill's theory for large-amplitude elongated bodies. For structural design parameters such as internal pressure,… ▽ More In this paper, we introduce the Soft Lamprey-Inspired Dual Environment Robot (SLIDER) and a proper modeling and optimization procedure employed to design the robot. We represent the primary fluid environment actions - inertial effects, vortex forces, and viscous dissipation - using Lighthill's theory for large-amplitude elongated bodies. For structural design parameters such as internal pressure, tail size, and body stiffness, a fast, geometrically and materially nonlinear model is developed and validated. The fluid-structure interaction equations are solved implicitly with an efficient second-order box method. A pneumatic manifold robotic system is employed to actuate SLIDER in a quiescent water tank environment, allowing cross-comparison of computational and experimental results. We find that low-frequency swimming is dominated by resistant environmental forces, whereas higher-frequency swimming is primarily affected by inertial fluid forces. Using our efficient model alongside a genetic algorithm, we co-optimize a swimming control pattern and caudal fin design (subject to SLIDER's climbing morphology) to achieve a tethered swimming speed of 21.7 +/- 0.4 cm/s (0.59 Bl/s). Furthermore, we investigate the optimization procedure for a multimodal robot performing both swimming and climbing tasks. △ Less

Submitted 9 June, 2026; originally announced June 2026.

Product units in gated recurrent units improve nuclear-mass prediction

Authors: Ziyuan Li, Paulo S. A. Freitas, John W. Clark, Babette Dellen

Abstract: The prediction of masses of atomic nuclei using machine learning can complement theoretical models and advance the exploration of poorly known domains of the nuclear chart. We propose a machine learning technique based on gated recurrent units (GRU), which have demonstrated competitive performance in nuclear-mass prediction by exploiting long-term dependencies. By integrating multiplicative intera… ▽ More The prediction of masses of atomic nuclei using machine learning can complement theoretical models and advance the exploration of poorly known domains of the nuclear chart. We propose a machine learning technique based on gated recurrent units (GRU), which have demonstrated competitive performance in nuclear-mass prediction by exploiting long-term dependencies. By integrating multiplicative interactions and product-unit transformations within recurrent units, we report significant improvements in nuclear-mass prediction. Computations are performed in the complex domain to jointly capture amplitude and phase dynamics. For interpolation and temporal-extrapolation tasks based on the atomic mass evaluation (AME2016 and AME2020), the complex additive-multiplicative product-unit gated recurrent unit (AM-PU-GRU) model consistently achieves the lowest prediction errors, with an interpolation RMSE of 0.227 $\pm$ 0.004 MeV and an extrapolation RMSE of 0.179 $\pm$ 0.015 MeV. These results surpass other state-of-the-art machine learning models and also outperform the real-valued GRU baseline and product-unit ablation variants, while remaining robust to different theoretical priors, including WS4 and SEMF. Our findings establish complex-valued product-unit recurrent networks as a new benchmark for sequence-based nuclear-mass prediction. △ Less

Submitted 4 June, 2026; originally announced June 2026.

Comments: Accepted at ICCS 2026

Journal ref: In: Lecture Notes in Computer Science, Springer, 2026

SREGym: A Live Benchmark for AI SRE Agents with High-Fidelity Failure Scenarios

Authors: Jackson Clark, Yiming Su, Saad Mohammad Rafid Pial, Yifang Tian, Lily Gniedziejko, Hans-Arno Jacobsen, Yinfang Chen, Tianyin Xu

Abstract: AI agents are increasingly used to diagnose and mitigate failures in production systems, known as agentic Site Reliability Engineering (SRE). Current SRE benchmarks are limited to oversimplistic SRE tasks and are unfortunately hard to extend due to bespoke designs. We present SREGym, a high-fidelity benchmark for SRE agents. SREGym exposes a live system environment built atop real-world cloud-nati… ▽ More AI agents are increasingly used to diagnose and mitigate failures in production systems, known as agentic Site Reliability Engineering (SRE). Current SRE benchmarks are limited to oversimplistic SRE tasks and are unfortunately hard to extend due to bespoke designs. We present SREGym, a high-fidelity benchmark for SRE agents. SREGym exposes a live system environment built atop real-world cloud-native system stacks, where high-fidelity failure scenarios are simulated through fault injectors. SREGym models the complexity of production environments by simulating (1) a wide range of faults at different layers, (2) various ambient noises, and (3) diverse failure modes such as metastable failures and correlated failures. SREGym is architected as a modular, extensible framework that orchestrates fault and noise injectors across stacks. SREGym currently includes 90 realistic, challenging SRE problems. We use SREGym to evaluate frontier agents and show that their capabilities varies significantly in addressing different kinds of failures, with up to 40% differences in end-to-end results. SREGym is actively maintained as an open-source project and has been used by researchers and practitioners. △ Less

Submitted 30 July, 2026; v1 submitted 7 May, 2026; originally announced May 2026.

Auditing Blockchain Innovations: Technical Challenges Beyond Traditional Finance

Authors: Shayan Eskandari, Leid Zejnilovic, Jeremy Clark

Abstract: Blockchain technology introduces asset types and custody mechanisms that fundamentally break traditional financial auditing paradigms. This paper presents an autoethnographic analysis of cryptoasset auditing challenges, build on top of prior research on a comprehensive framework addressing existence, ownership, valuation, and internal control verification. Drawing from lived experience implementin… ▽ More Blockchain technology introduces asset types and custody mechanisms that fundamentally break traditional financial auditing paradigms. This paper presents an autoethnographic analysis of cryptoasset auditing challenges, build on top of prior research on a comprehensive framework addressing existence, ownership, valuation, and internal control verification. Drawing from lived experience implementing blockchain systems as an engineer, smart contract auditor, and CTO of a publicly traded cryptoasset firm, we demonstrate how autoethnographic methodology becomes necessary for understanding technical complexities that external analysis cannot capture. Through detailed examination of token airdrops, multi-signature smart contracts, and real-time on-chain reporting, we provide experimental approaches and common scenarios that auditing firms can analyze to address blockchain innovations currently considered technically insurmountable. △ Less

Submitted 27 March, 2026; originally announced March 2026.

Comments: 6 pages, short paper, 4 figures, Blockchain Confluence, IEEE International Conference on Distributed Ledger Technologies

Responsible Intelligence in Practice: A Fairness Audit of Open Large Language Models for Library Reference Services

Authors: Haining Wang, Jason Clark, Angelica Peña

Abstract: As libraries explore large language models (LLMs) as a scalable layer for reference services, a core fairness question follows: can LLM-based services support all patrons fairly, regardless of demographic identity? While LLMs offer great potential for broadening access to information assistance, they may also reproduce societal biases embedded in their training data, potentially undermining librar… ▽ More As libraries explore large language models (LLMs) as a scalable layer for reference services, a core fairness question follows: can LLM-based services support all patrons fairly, regardless of demographic identity? While LLMs offer great potential for broadening access to information assistance, they may also reproduce societal biases embedded in their training data, potentially undermining libraries' commitments to impartial service. In this chapter, we apply a systematic evaluation approach that combines diagnostic classification to detect systematic differences with linguistic analysis to interpret their sources. Across three widely used open models (Llama-3.1 8B, Gemma-2 9B, and Ministral 8B), we find no compelling evidence of systematic differentiation by race/ethnicity, and only minor evidence of sex-linked differentiation in one model. We discuss implications for responsible AI adoption in libraries and the importance of ongoing monitoring in aligning LLM-based services with core professional values. △ Less

Submitted 21 February, 2026; originally announced February 2026.

Comments: Invited chapter for the edited volume Artificial Intelligence and Social Justice Intersections in Library and Information Studies: Challenges and Opportunities (Emerald Group Publishing, in preparation)

In-Network Collective Operations: Game Changer or Challenge for AI Workloads?

Authors: Torsten Hoefler, Mikhail Khalilov, Josiah Clark, Surendra Anubolu, Mohan Kalkunte, Karen Schramm, Eric Spada, Duncan Roweth, Keith Underwood, Adrian Caulfield, Abdul Kabbani, Amirreza Rastegari

Abstract: This paper summarizes the opportunities of in-network collective operations (INC) for accelerated collective operations in AI workloads. We provide sufficient detail to make this important field accessible to non-experts in AI or networking, fostering a connection between these communities. Consider two types of INC: Edge-INC, where the system is implemented at the node level, and Core-INC, where… ▽ More This paper summarizes the opportunities of in-network collective operations (INC) for accelerated collective operations in AI workloads. We provide sufficient detail to make this important field accessible to non-experts in AI or networking, fostering a connection between these communities. Consider two types of INC: Edge-INC, where the system is implemented at the node level, and Core-INC, where the system is embedded within network switches. We outline the potential performance benefits as well as six key obstacles in the context of both Edge-INC and Core-INC that may hinder their adoption. Finally, we present a set of predictions for the future development and application of INC. △ Less

Submitted 26 January, 2026; originally announced January 2026.

Journal ref: IEEE Computer Jan. 2026

Ego4OOD: Rethinking Egocentric Video Domain Generalization via Covariate Shift Scoring

Authors: Zahra Vaseqi, James Clark

Abstract: Egocentric video action recognition under domain shifts remains challenging due to large intra-class spatio-temporal variability, long-tailed feature distributions, and strong correlations between actions and environments. Existing benchmarks for egocentric domain generalization often conflate covariate shifts with concept shifts, making it difficult to reliably evaluate a model's ability to gener… ▽ More Egocentric video action recognition under domain shifts remains challenging due to large intra-class spatio-temporal variability, long-tailed feature distributions, and strong correlations between actions and environments. Existing benchmarks for egocentric domain generalization often conflate covariate shifts with concept shifts, making it difficult to reliably evaluate a model's ability to generalize across input distributions. To address this limitation, we introduce Ego4OOD, a domain generalization benchmark derived from Ego4D that emphasizes measurable covariate diversity while reducing concept shift through semantically coherent, moment-level action categories. Ego4OOD spans eight geographically distinct domains and is accompanied by a clustering-based covariate shift metric that provides a quantitative proxy for domain difficulty. We further leverage a one-vs-all binary training objective that decomposes multi-class action recognition into independent binary classification tasks. This formulation is particularly well-suited for covariate shift by reducing interference between visually similar classes under feature distribution shift. Using this formulation, we show that a lightweight two-layer fully connected network achieves performance competitive with state-of-the-art egocentric domain generalization methods on both Argo1M and Ego4OOD, despite using fewer parameters and no additional modalities. Our empirical analysis demonstrates a clear relationship between measured covariate shift and recognition performance, highlighting the importance of controlled benchmarks and quantitative domain characterization for studying out-of-distribution generalization in egocentric video. △ Less

Submitted 21 January, 2026; originally announced January 2026.

AntiPaSTO: Self-Supervised Honesty Steering via Anti-Parallel Representations

Authors: Michael J. Clark

Abstract: As models grow more capable, humans cannot reliably verify what they say. Scalable steering requires methods that are internal, self-supervised, and transfer out-of-distribution; existing methods satisfy some but not all three. We introduce AntiPaSTO, which separates representations along an antiparallel axis (+1/-1 produce opposite shifts), with coherence constraints preventing collapse. Training… ▽ More As models grow more capable, humans cannot reliably verify what they say. Scalable steering requires methods that are internal, self-supervised, and transfer out-of-distribution; existing methods satisfy some but not all three. We introduce AntiPaSTO, which separates representations along an antiparallel axis (+1/-1 produce opposite shifts), with coherence constraints preventing collapse. Training uses only two contrasting words inserted into template sentences, with no preference labels. When we use 800 such synthetic pairs on Gemma-3-1B, AntiPaSTO beats prompting baselines by 6.9x Steering F1 on DailyDilemmas and wins on 5 of 6 tested value axes. We also find preliminary evidence that it maintains bidirectional control where prompting triggers refusal. △ Less

Submitted 11 May, 2026; v1 submitted 12 January, 2026; originally announced January 2026.

Comments: Code is available at https://github.com/wassname/AntiPaSTO

NEMO-4-PAYPAL: Leveraging NVIDIA's Nemo Framework for empowering PayPal's Commerce Agent

Authors: Sudhanshu Garg, Andrew Wang, Chaitanya Kulkarni, Ali Sahami, Farhad Farahani, Sean Yun-Shiuan Chuang, Jian Wan, Srinivasan Manoharan, Uma Kona, Nitin Sharma, Linsey Pang, Prakhar Mehrotra, Jessica Clark, Mark Moyou

Abstract: We present the development and optimization of PayPal's Commerce Agent, powered by NEMO-4-PAYPAL, a multi-agent system designed to revolutionize agentic commerce on the PayPal platform. Through our strategic partnership with NVIDIA, we leveraged the NeMo Framework for LLM model fine-tuning to enhance agent performance. Specifically, we optimized the Search and Discovery agent by replacing our base… ▽ More We present the development and optimization of PayPal's Commerce Agent, powered by NEMO-4-PAYPAL, a multi-agent system designed to revolutionize agentic commerce on the PayPal platform. Through our strategic partnership with NVIDIA, we leveraged the NeMo Framework for LLM model fine-tuning to enhance agent performance. Specifically, we optimized the Search and Discovery agent by replacing our base model with a fine-tuned Nemotron small language model (SLM). We conducted comprehensive experiments using the llama3.1-nemotron-nano-8B-v1 architecture, training LoRA-based models through systematic hyperparameter sweeps across learning rates, optimizers (Adam, AdamW), cosine annealing schedules, and LoRA ranks. Our contributions include: (1) the first application of NVIDIA's NeMo Framework to commerce-specific agent optimization, (2) LLM powered fine-tuning strategy for retrieval-focused commerce tasks, (3) demonstration of significant improvements in latency and cost while maintaining agent quality, and (4) a scalable framework for multi-agent system optimization in production e-commerce environments. Our results demonstrate that the fine-tuned Nemotron SLM effectively resolves the key performance issue in the retrieval component, which represents over 50\% of total agent response time, while maintaining or enhancing overall system performance. △ Less

Submitted 7 January, 2026; v1 submitted 25 December, 2025; originally announced December 2025.

FLEX: Feature Importance from Layered Counterfactual Explanations

Authors: Nawid Keshtmand, Roussel Desmond Nzoyem, Jeffrey Nicholas Clark

Abstract: Machine learning models achieve state-of-the-art performance across domains, yet their lack of interpretability limits safe deployment in high-stakes settings. Counterfactual explanations are widely used to provide actionable "what-if" recourse, but they typically remain instance-specific and do not quantify which features systematically drive outcome changes within coherent regions of the feature… ▽ More Machine learning models achieve state-of-the-art performance across domains, yet their lack of interpretability limits safe deployment in high-stakes settings. Counterfactual explanations are widely used to provide actionable "what-if" recourse, but they typically remain instance-specific and do not quantify which features systematically drive outcome changes within coherent regions of the feature space or across an entire dataset. We introduce FLEX (Feature importance from Layered counterfactual EXplanations), a model- and domain-agnostic framework that converts sets of counterfactuals into feature change frequency scores at local, regional, and global levels. FLEX generalises local change-frequency measures by aggregating across instances and neighbourhoods, offering interpretable rankings that reflect how often each feature must change to flip predictions. The framework is compatible with different counterfactual generation methods, allowing users to emphasise characteristics such as sparsity, feasibility, or actionability, thereby tailoring the derived feature importances to practical constraints. We evaluate FLEX on two contrasting tabular tasks: traffic accident severity prediction and loan approval, and compare FLEX to SHAP- and LIME-derived feature importance values. Results show that (i) FLEX's global rankings correlate with SHAP while surfacing additional drivers, and (ii) regional analyses reveal context-specific factors that global summaries miss. FLEX thus bridges the gap between local recourse and global attribution, supporting transparent and intervention-oriented decision-making in risk-sensitive applications. △ Less

Submitted 14 November, 2025; originally announced November 2025.

Comments: 12 pages, 6 figures, 3 tables, 2 algorithms. Preprint under review

Machine Learning for Climate Policy: Understanding Policy Progression in the European Green Deal

Authors: Patricia West, Michelle WL Wan, Alexander Hepburn, Edwin Simpson, Raul Santos-Rodriguez, Jeffrey N Clark

Abstract: Climate change demands effective legislative action to mitigate its impacts. This study explores the application of machine learning (ML) to understand the progression of climate policy from announcement to adoption, focusing on policies within the European Green Deal. We present a dataset of 165 policies, incorporating text and metadata. We aim to predict a policy's progression status, and compar… ▽ More Climate change demands effective legislative action to mitigate its impacts. This study explores the application of machine learning (ML) to understand the progression of climate policy from announcement to adoption, focusing on policies within the European Green Deal. We present a dataset of 165 policies, incorporating text and metadata. We aim to predict a policy's progression status, and compare text representation methods, including TF-IDF, BERT, and ClimateBERT. Metadata features are included to evaluate the impact on predictive performance. On text features alone, ClimateBERT outperforms other approaches (RMSE = 0.17, R^2 = 0.29), while BERT achieves superior performance with the addition of metadata features (RMSE = 0.16, R^2 = 0.38). Using methods from explainable AI highlights the influence of factors such as policy wording and metadata including political party and country representation. These findings underscore the potential of ML tools in supporting climate policy analysis and decision-making. △ Less

Submitted 17 October, 2025; originally announced October 2025.

SoK: Market Microstructure for Decentralized Prediction Markets (DePMs)

Authors: Nahid Rahman, Joseph Al-Chami, Jeremy Clark

Abstract: Decentralized prediction markets (DePMs) allow open participation in event-based wagering without fully relying on centralized intermediaries. We review the history of DePMs which date back to 2011 and includes hundreds of proposals. Perhaps surprising, modern DePMs like Polymarket deviate materially from earlier designs like Truthcoin and Augur v1. We use our review to present a modular workflow… ▽ More Decentralized prediction markets (DePMs) allow open participation in event-based wagering without fully relying on centralized intermediaries. We review the history of DePMs which date back to 2011 and includes hundreds of proposals. Perhaps surprising, modern DePMs like Polymarket deviate materially from earlier designs like Truthcoin and Augur v1. We use our review to present a modular workflow comprising eight stages: underlying infrastructure, market topic, share structure and pricing, market initialization, trading, market resolution, settlement, and archiving. For each module, we enumerate the design variants, analyzing trade-offs around decentralization, expressiveness, and manipulation resistance. We also identify open problems for researchers interested in this ecosystem. △ Less

Submitted 8 July, 2026; v1 submitted 17 October, 2025; originally announced October 2025.

Uncertainty assessment in satellite-based greenhouse gas emissions estimates using emulated atmospheric transport

Authors: Jeffrey N. Clark, Elena Fillola, Nawid Keshtmand, Raul Santos-Rodriguez, Matthew Rigby

Abstract: Monitoring greenhouse gas emissions and evaluating national inventories require efficient, scalable, and reliable inference methods. Top-down approaches, combined with recent advances in satellite observations, provide new opportunities to evaluate emissions at continental and global scales. However, transport models used in these methods remain a key source of uncertainty: they are computationall… ▽ More Monitoring greenhouse gas emissions and evaluating national inventories require efficient, scalable, and reliable inference methods. Top-down approaches, combined with recent advances in satellite observations, provide new opportunities to evaluate emissions at continental and global scales. However, transport models used in these methods remain a key source of uncertainty: they are computationally expensive to run at scale, and their uncertainty is difficult to characterise. Artificial intelligence offers a dual opportunity to accelerate transport simulations and to quantify their associated uncertainty. We present an ensemble-based pipeline for estimating atmospheric transport "footprints", greenhouse gas mole fraction measurements, and their uncertainties using a graph neural network emulator of a Lagrangian Particle Dispersion Model (LPDM). The approach is demonstrated with GOSAT (Greenhouse Gases Observing Satellite) observations for Brazil in 2016. The emulator achieved a ~1000x speed-up over the NAME LPDM, while reproducing large-scale footprint structures. Ensembles were calculated to quantify absolute and relative uncertainty, revealing spatial correlations with prediction error. The results show that ensemble spread highlights low-confidence spatial and temporal predictions for both atmospheric transport footprints and methane mole fractions. While demonstrated here for an LPDM emulator, the approach could be applied more generally to atmospheric transport models, supporting uncertainty-aware greenhouse gas inversion systems and improving the robustness of satellite-based emissions monitoring. With further development, ensemble-based emulators could also help explore systematic LPDM errors, offering a computationally efficient pathway towards a more comprehensive uncertainty budget in greenhouse gas flux estimates. △ Less

Submitted 7 October, 2025; originally announced October 2025.

Model-agnostic Meta-learning for Adaptive Gait Phase and Terrain Geometry Estimation with Wearable Soft Sensors

Authors: Zenan Zhu, Wenxi Chen, Pei-Chun Kao, Janelle Clark, Lily Behnke, Rebecca Kramer-Bottiglio, Holly Yanco, Yan Gu

Abstract: This letter presents a model-agnostic meta-learning (MAML) based framework for simultaneous and accurate estimation of human gait phase and terrain geometry using a small set of fabric-based wearable soft sensors, with efficient adaptation to unseen subjects and strong generalization across different subjects and terrains. Compared to rigid alternatives such as inertial measurement units, fabric-b… ▽ More This letter presents a model-agnostic meta-learning (MAML) based framework for simultaneous and accurate estimation of human gait phase and terrain geometry using a small set of fabric-based wearable soft sensors, with efficient adaptation to unseen subjects and strong generalization across different subjects and terrains. Compared to rigid alternatives such as inertial measurement units, fabric-based soft sensors improve comfort but introduce nonlinearities due to hysteresis, placement error, and fabric deformation. Moreover, inter-subject and inter-terrain variability, coupled with limited calibration data in real-world deployments, further complicate accurate estimation. To address these challenges, the proposed framework integrates MAML into a deep learning architecture to learn a generalizable model initialization that captures subject- and terrain-invariant structure. This initialization enables efficient adaptation (i.e., adaptation with only a small amount of calibration data and a few fine-tuning steps) to new users, while maintaining strong generalization (i.e., high estimation accuracy across subjects and terrains). Experiments on nine participants walking at various speeds over five terrain conditions demonstrate that the proposed framework outperforms baseline approaches in estimating gait phase, locomotion mode, and incline angle, with superior accuracy, adaptation efficiency, and generalization. △ Less

Submitted 4 August, 2025; originally announced August 2025.

Comments: 8 pages, 5 figures

Improving Community Detection in Academic Networks by Handling Publication Bias

Authors: Md Asaduzzaman Noor, John Sheppard, Jason Clark

Abstract: Finding potential research collaborators is a challenging task, especially in today's fast-growing and interdisciplinary research landscape. While traditional methods often rely on observable relationships such as co-authorships and citations to construct the research network, in this work, we focus solely on publication content to build a topic-based research network using BERTopic with a fine-tu… ▽ More Finding potential research collaborators is a challenging task, especially in today's fast-growing and interdisciplinary research landscape. While traditional methods often rely on observable relationships such as co-authorships and citations to construct the research network, in this work, we focus solely on publication content to build a topic-based research network using BERTopic with a fine-tuned SciBERT model that connects and recommends researchers across disciplines based on shared topical interests. A major challenge we address is publication imbalance, where some researchers publish much more than others, often across several topics. Without careful handling, their less frequent interests are hidden under dominant topics, limiting the network's ability to detect their full research scope. To tackle this, we introduce a cloning strategy that clusters a researcher's publications and treats each cluster as a separate node. This allows researchers to be part of multiple communities, improving the detection of interdisciplinary links. Evaluation on the proposed method shows that the cloned network structure leads to more meaningful communities and uncovers a broader set of collaboration opportunities. △ Less

Submitted 27 July, 2025; originally announced July 2025.

Comments: This paper is an extended version of a work accepted at ASONAM 2025

TyDi QA-WANA: A Benchmark for Information-Seeking Question Answering in Languages of West Asia and North Africa

Authors: Parker Riley, Siamak Shakeri, Waleed Ammar, Jonathan H. Clark

Abstract: We present TyDi QA-WANA, a question-answering dataset consisting of 28K examples divided among 10 language varieties of western Asia and northern Africa. The data collection process was designed to elicit information-seeking questions, where the asker is genuinely curious to know the answer. Each question in paired with an entire article that may or may not contain the answer; the relatively large… ▽ More We present TyDi QA-WANA, a question-answering dataset consisting of 28K examples divided among 10 language varieties of western Asia and northern Africa. The data collection process was designed to elicit information-seeking questions, where the asker is genuinely curious to know the answer. Each question in paired with an entire article that may or may not contain the answer; the relatively large size of the articles results in a task suitable for evaluating models' abilities to utilize large text contexts in answering questions. Furthermore, the data was collected directly in each language variety, without the use of translation, in order to avoid issues of cultural relevance. We present performance of two baseline models, and release our code and data to facilitate further improvement by the research community. △ Less

Submitted 23 July, 2025; originally announced July 2025.

Event Detection for Active Lower Limb Prosthesis

Authors: J. D. Clark, P. Ellison

Abstract: Accurate event detection is key to the successful design of semi-passive and powered prosthetics. Kinematically, the natural knee is complex, with translation and rotation components that have a substantial impact on gait characteristics. When simplified to a pin joint, some of this behaviour is lost. This study investigates the role of cruciate ligament stretch in event detection. A bicondylar kn… ▽ More Accurate event detection is key to the successful design of semi-passive and powered prosthetics. Kinematically, the natural knee is complex, with translation and rotation components that have a substantial impact on gait characteristics. When simplified to a pin joint, some of this behaviour is lost. This study investigates the role of cruciate ligament stretch in event detection. A bicondylar knee design was used, constrained by analogues of the anterior and posterior cruciate ligaments. This offers the ability to characterize knee kinematics by the stretch of the ligaments. The ligament stretch was recorded using LVDTs parallel to the ligaments of the Russell knee on a bent knee crutch. Which was used to capture data on a treadmill at 3 speeds. This study finds speed dependence within the stretch of the cruciate ligaments, prominently around 5\% and 80\% of the gait cycle for the posterior and anterior. The cycle profile remains consistent with speed; therefore, other static events such as the turning point feature at around 90\% and 95\% of the cycle, for the posterior and anterior, respectively, could be used as a predictive precursor for initial contact. Likewise at 90\% and 95\%, another pair of turning points that in this case could be used to predict foot flat. This concludes that the use of a bicondylar knee design could improve the detection of events during the gait cycle, and therefore could increase the accuracy of subsequent controllers for powered prosthetics. △ Less

Submitted 23 July, 2025; originally announced July 2025.

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Authors: Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, Krishna Haridasan, Ahmed Omran, Nikunj Saunshi, Dara Bahri, Gaurav Mishra, Eric Chu , et al. (3410 additional authors not shown)

Abstract: In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal unde… ▽ More In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving. △ Less

Submitted 19 December, 2025; v1 submitted 7 July, 2025; originally announced July 2025.

Comments: 72 pages, 17 figures

Fairness Evaluation of Large Language Models in Academic Library Reference Services

Authors: Haining Wang, Jason Clark, Yueru Yan, Star Bradley, Ruiyang Chen, Yiqiong Zhang, Hengyi Fu, Zuoyu Tian

Abstract: As libraries explore large language models (LLMs) for use in virtual reference services, a key question arises: Can LLMs serve all users equitably, regardless of demographics or social status? While they offer great potential for scalable support, LLMs may also reproduce societal biases embedded in their training data, risking the integrity of libraries' commitment to equitable service. To address… ▽ More As libraries explore large language models (LLMs) for use in virtual reference services, a key question arises: Can LLMs serve all users equitably, regardless of demographics or social status? While they offer great potential for scalable support, LLMs may also reproduce societal biases embedded in their training data, risking the integrity of libraries' commitment to equitable service. To address this concern, we evaluate whether LLMs differentiate responses across user identities by prompting six state-of-the-art LLMs to assist patrons differing in sex, race/ethnicity, and institutional role. We find no evidence of differentiation by race or ethnicity, and only minor evidence of stereotypical bias against women in one model. LLMs demonstrate nuanced accommodation of institutional roles through the use of linguistic choices related to formality, politeness, and domain-specific vocabularies, reflecting professional norms rather than discriminatory treatment. These findings suggest that current LLMs show a promising degree of readiness to support equitable and contextually appropriate communication in academic library reference services. △ Less

Submitted 21 November, 2025; v1 submitted 5 July, 2025; originally announced July 2025.

STRATUS: A Multi-agent System for Autonomous Reliability Engineering of Modern Clouds

Authors: Yinfang Chen, Jiaqi Pan, Jackson Clark, Yiming Su, Noah Zheutlin, Bhavya Bhavya, Rohan Arora, Yu Deng, Saurabh Jha, Tianyin Xu

Abstract: In cloud-scale systems, failures are the norm. A distributed computing cluster exhibits hundreds of machine failures and thousands of disk failures; software bugs and misconfigurations are reported to be more frequent. The demand for autonomous, AI-driven reliability engineering continues to grow, as existing humanin-the-loop practices can hardly keep up with the scale of modern clouds. This paper… ▽ More In cloud-scale systems, failures are the norm. A distributed computing cluster exhibits hundreds of machine failures and thousands of disk failures; software bugs and misconfigurations are reported to be more frequent. The demand for autonomous, AI-driven reliability engineering continues to grow, as existing humanin-the-loop practices can hardly keep up with the scale of modern clouds. This paper presents STRATUS, an LLM-based multi-agent system for realizing autonomous Site Reliability Engineering (SRE) of cloud services. STRATUS consists of multiple specialized agents (e.g., for failure detection, diagnosis, mitigation), organized in a state machine to assist system-level safety reasoning and enforcement. We formalize a key safety specification of agentic SRE systems like STRATUS, termed Transactional No-Regression (TNR), which enables safe exploration and iteration. We show that TNR can effectively improve autonomous failure mitigation. STRATUS significantly outperforms state-of-the-art SRE agents in terms of success rate of failure mitigation problems in AIOpsLab and ITBench (two SRE benchmark suites), by at least 1.5 times across various models. STRATUS shows a promising path toward practical deployment of agentic systems for cloud reliability. △ Less

Submitted 19 March, 2026; v1 submitted 27 May, 2025; originally announced June 2025.

Comments: 10 pages for main text

Improving Local Air Quality Predictions Using Transfer Learning on Satellite Data and Graph Neural Networks

Authors: Finn Gueterbock, Raul Santos-Rodriguez, Jeffrey N. Clark

Abstract: Air pollution is a significant global health risk, contributing to millions of premature deaths annually. Nitrogen dioxide (NO2), a harmful pollutant, disproportionately affects urban areas where monitoring networks are often sparse. We propose a novel method for predicting NO2 concentrations at unmonitored locations using transfer learning with satellite and meteorological data. Leveraging the Gr… ▽ More Air pollution is a significant global health risk, contributing to millions of premature deaths annually. Nitrogen dioxide (NO2), a harmful pollutant, disproportionately affects urban areas where monitoring networks are often sparse. We propose a novel method for predicting NO2 concentrations at unmonitored locations using transfer learning with satellite and meteorological data. Leveraging the GraphSAGE framework, our approach integrates autoregression and transfer learning to enhance predictive accuracy in data-scarce regions like Bristol. Pre-trained on data from London, UK, our model achieves a 8.6% reduction in Normalised Root Mean Squared Error (NRMSE) and a 32.6% reduction in Gradient RMSE compared to a baseline model. This work demonstrates the potential of virtual sensors for cost-effective air quality monitoring, contributing to actionable insights for climate and health interventions. △ Less

Submitted 23 April, 2025; originally announced May 2025.

Prototype-enhanced prediction in graph neural networks for climate applications

Authors: Nawid Keshtmand, Elena Fillola, Jeffrey Nicholas Clark, Raul Santos-Rodriguez, Matthew Rigby

Abstract: Data-driven emulators are increasingly being used to learn and emulate physics-based simulations, reducing computational expense and run time. Here, we present a structured way to improve the quality of these high-dimensional emulated outputs, through the use of prototypes: an approximation of the emulator's output passed as an input, which informs the model and leads to better predictions. We dem… ▽ More Data-driven emulators are increasingly being used to learn and emulate physics-based simulations, reducing computational expense and run time. Here, we present a structured way to improve the quality of these high-dimensional emulated outputs, through the use of prototypes: an approximation of the emulator's output passed as an input, which informs the model and leads to better predictions. We demonstrate our approach to emulate atmospheric dispersion, key for greenhouse gas emissions monitoring, by comparing a baseline model to models trained using prototypes as an additional input. The prototype models achieve better performance, even with few prototypes and even if they are chosen at random, but we show that choosing the prototypes through data-driven methods (k-means) can lead to almost 10\% increased performance in some metrics. △ Less

Submitted 24 April, 2025; originally announced April 2025.

Neural Radiance and Gaze Fields for Visual Attention Modeling in 3D Environments

Authors: Andrei Chubarau, Yinan Wang, James J. Clark

Abstract: We introduce Neural Radiance and Gaze Fields (NeRGs), a novel approach for representing visual attention in complex environments. Much like how Neural Radiance Fields (NeRFs) perform novel view synthesis, NeRGs reconstruct gaze patterns from arbitrary viewpoints, implicitly mapping visual attention to 3D surfaces. We achieve this by augmenting a standard NeRF with an additional network that models… ▽ More We introduce Neural Radiance and Gaze Fields (NeRGs), a novel approach for representing visual attention in complex environments. Much like how Neural Radiance Fields (NeRFs) perform novel view synthesis, NeRGs reconstruct gaze patterns from arbitrary viewpoints, implicitly mapping visual attention to 3D surfaces. We achieve this by augmenting a standard NeRF with an additional network that models local egocentric gaze probability density, conditioned on scene geometry and observer position. The output of a NeRG is a rendered view of the scene alongside a pixel-wise salience map representing the conditional probability that a given observer fixates on visible surfaces. Unlike prior methods, our system is lightweight and enables visualization of gaze fields at interactive framerates. Moreover, NeRGs allow the observer perspective to be decoupled from the rendering camera and correctly account for gaze occlusion due to intervening geometry. We demonstrate the effectiveness of NeRGs using head pose from skeleton tracking as a proxy for gaze, employing our proposed gaze probes to aggregate noisy rays into robust probability density targets for supervision. △ Less

Submitted 3 December, 2025; v1 submitted 10 March, 2025; originally announced March 2025.

Comments: 11 pages, 8 figures

Which Economic Tasks are Performed with AI? Evidence from Millions of Claude Conversations

Authors: Kunal Handa, Alex Tamkin, Miles McCain, Saffron Huang, Esin Durmus, Sarah Heck, Jared Mueller, Jerry Hong, Stuart Ritchie, Tim Belonax, Kevin K. Troy, Dario Amodei, Jared Kaplan, Jack Clark, Deep Ganguli

Abstract: Despite widespread speculation about artificial intelligence's impact on the future of work, we lack systematic empirical evidence about how these systems are actually being used for different tasks. Here, we present a novel framework for measuring AI usage patterns across the economy. We leverage a recent privacy-preserving system to analyze over four million Claude.ai conversations through the l… ▽ More Despite widespread speculation about artificial intelligence's impact on the future of work, we lack systematic empirical evidence about how these systems are actually being used for different tasks. Here, we present a novel framework for measuring AI usage patterns across the economy. We leverage a recent privacy-preserving system to analyze over four million Claude.ai conversations through the lens of tasks and occupations in the U.S. Department of Labor's O*NET Database. Our analysis reveals that AI usage primarily concentrates in software development and writing tasks, which together account for nearly half of all total usage. However, usage of AI extends more broadly across the economy, with approximately 36% of occupations using AI for at least a quarter of their associated tasks. We also analyze how AI is being used for tasks, finding 57% of usage suggests augmentation of human capabilities (e.g., learning or iterating on an output) while 43% suggests automation (e.g., fulfilling a request with minimal human involvement). While our data and methods face important limitations and only paint a picture of AI usage on a single platform, they provide an automated, granular approach for tracking AI's evolving role in the economy and identifying leading indicators of future impact as these technologies continue to advance. △ Less

Submitted 10 February, 2025; originally announced March 2025.

Performance Evaluation of Large Language Models in Statistical Programming

Authors: Xinyi Song, Kexin Xie, Lina Lee, Ruizhe Chen, Jared M. Clark, Hao He, Haoran He, Jie Min, Xinlei Zhang, Simin Zheng, Zhiyang Zhang, Xinwei Deng, Yili Hong

Abstract: The programming capabilities of large language models (LLMs) have revolutionized automatic code generation and opened new avenues for automatic statistical analysis. However, the validity and quality of these generated codes need to be systematically evaluated before they can be widely adopted. Despite their growing prominence, a comprehensive evaluation of statistical code generated by LLMs remai… ▽ More The programming capabilities of large language models (LLMs) have revolutionized automatic code generation and opened new avenues for automatic statistical analysis. However, the validity and quality of these generated codes need to be systematically evaluated before they can be widely adopted. Despite their growing prominence, a comprehensive evaluation of statistical code generated by LLMs remains scarce in the literature. In this paper, we assess the performance of LLMs, including two versions of ChatGPT and one version of Llama, in the domain of SAS programming for statistical analysis. Our study utilizes a set of statistical analysis tasks encompassing diverse statistical topics and datasets. Each task includes a problem description, dataset information, and human-verified SAS code. We conduct a comprehensive assessment of the quality of SAS code generated by LLMs through human expert evaluation based on correctness, effectiveness, readability, executability, and the accuracy of output results. The analysis of rating scores reveals that while LLMs demonstrate usefulness in generating syntactically correct code, they struggle with tasks requiring deep domain understanding and may produce redundant or incorrect results. This study offers valuable insights into the capabilities and limitations of LLMs in statistical programming, providing guidance for future advancements in AI-assisted coding systems for statistical analysis. △ Less

Submitted 18 February, 2025; originally announced February 2025.

Comments: 27 pages, 8 figures

Bridging the Data Gap in AI Reliability Research and Establishing DR-AIR, a Comprehensive Data Repository for AI Reliability

Authors: Simin Zheng, Jared M. Clark, Fatemeh Salboukh, Priscila Silva, Karen da Mata, Fenglian Pan, Jie Min, Jiayi Lian, Caleb B. King, Lance Fiondella, Jian Liu, Xinwei Deng, Yili Hong

Abstract: Artificial intelligence (AI) technology and systems have been advancing rapidly. However, ensuring the reliability of these systems is crucial for fostering public confidence in their use. This necessitates the modeling and analysis of reliability data specific to AI systems. A major challenge in AI reliability research, particularly for those in academia, is the lack of readily available AI relia… ▽ More Artificial intelligence (AI) technology and systems have been advancing rapidly. However, ensuring the reliability of these systems is crucial for fostering public confidence in their use. This necessitates the modeling and analysis of reliability data specific to AI systems. A major challenge in AI reliability research, particularly for those in academia, is the lack of readily available AI reliability data. To address this gap, this paper focuses on conducting a comprehensive review of available AI reliability data and establishing DR-AIR: a data repository for AI reliability. Specifically, we introduce key measurements and data types for assessing AI reliability, along with the methodologies used to collect these data. We also provide a detailed description of the currently available datasets with illustrative examples. Furthermore, we outline the setup of the DR-AIR repository and demonstrate its practical applications. This repository provides easy access to datasets specifically curated for AI reliability research. We believe these efforts will significantly benefit the AI research community by facilitating access to valuable reliability data and promoting collaboration across various academic domains within AI. We conclude our paper with a call to action, encouraging the research community to contribute and share AI reliability data to further advance this critical field of study. △ Less

Submitted 17 February, 2025; originally announced February 2025.

Comments: 34 pages, 12 figures

ITBench: Evaluating AI Agents across Diverse Real-World IT Automation Tasks

Authors: Saurabh Jha, Rohan Arora, Yuji Watanabe, Takumi Yanagawa, Yinfang Chen, Jackson Clark, Bhavya Bhavya, Mudit Verma, Harshit Kumar, Hirokuni Kitahara, Noah Zheutlin, Saki Takano, Divya Pathak, Felix George, Xinbo Wu, Bekir O. Turkkan, Gerard Vanloo, Michael Nidd, Ting Dai, Oishik Chatterjee, Pranjal Gupta, Suranjana Samanta, Pooja Aggarwal, Rong Lee, Pavankumar Murali , et al. (18 additional authors not shown)

Abstract: Realizing the vision of using AI agents to automate critical IT tasks depends on the ability to measure and understand effectiveness of proposed solutions. We introduce ITBench, a framework that offers a systematic methodology for benchmarking AI agents to address real-world IT automation tasks. Our initial release targets three key areas: Site Reliability Engineering (SRE), Compliance and Securit… ▽ More Realizing the vision of using AI agents to automate critical IT tasks depends on the ability to measure and understand effectiveness of proposed solutions. We introduce ITBench, a framework that offers a systematic methodology for benchmarking AI agents to address real-world IT automation tasks. Our initial release targets three key areas: Site Reliability Engineering (SRE), Compliance and Security Operations (CISO), and Financial Operations (FinOps). The design enables AI researchers to understand the challenges and opportunities of AI agents for IT automation with push-button workflows and interpretable metrics. ITBench includes an initial set of 94 real-world scenarios, which can be easily extended by community contributions. Our results show that agents powered by state-of-the-art models resolve only 13.8% of SRE scenarios, 25.2% of CISO scenarios, and 0% of FinOps scenarios. We expect ITBench to be a key enabler of AI-driven IT automation that is correct, safe, and fast. △ Less

Submitted 7 February, 2025; originally announced February 2025.

Action-Free Reasoning for Policy Generalization

Authors: Jaden Clark, Suvir Mirchandani, Dorsa Sadigh, Suneel Belkhale

Abstract: End-to-end imitation learning offers a promising approach for training robot policies. However, generalizing to new settings remains a significant challenge. Although large-scale robot demonstration datasets have shown potential for inducing generalization, they are resource-intensive to scale. In contrast, human video data is abundant and diverse, presenting an attractive alternative. Yet, these… ▽ More End-to-end imitation learning offers a promising approach for training robot policies. However, generalizing to new settings remains a significant challenge. Although large-scale robot demonstration datasets have shown potential for inducing generalization, they are resource-intensive to scale. In contrast, human video data is abundant and diverse, presenting an attractive alternative. Yet, these human-video datasets lack action labels, complicating their use in imitation learning. Existing methods attempt to extract grounded action representations (e.g., hand poses), but resulting policies struggle to bridge the embodiment gap between human and robot actions. We propose an alternative approach: leveraging language-based reasoning from human videos-essential for guiding robot actions-to train generalizable robot policies. Building on recent advances in reasoning-based policy architectures, we introduce Reasoning through Action-free Data (RAD). RAD learns from both robot demonstration data (with reasoning and action labels) and action-free human video data (with only reasoning labels). The robot data teaches the model to map reasoning to low-level actions, while the action-free data enhances reasoning capabilities. Additionally, we will release a new dataset of 3,377 human-hand demonstrations with reasoning annotations compatible with the Bridge V2 benchmark and aimed at facilitating future research on reasoning-driven robot learning. Our experiments show that RAD enables effective transfer across the embodiment gap, allowing robots to perform tasks seen only in action-free data. Furthermore, scaling up action-free reasoning data significantly improves policy performance and generalization to novel tasks. These results highlight the promise of reasoning-driven learning from action-free datasets for advancing generalizable robot control. Project page: https://rad-generalization.github.io △ Less

Submitted 10 February, 2025; v1 submitted 5 February, 2025; originally announced February 2025.

Comments: 13 pages, 10 figures

Efficiently Generating Expressive Quadruped Behaviors via Language-Guided Preference Learning

Authors: Jaden Clark, Joey Hejna, Dorsa Sadigh

Abstract: Expressive robotic behavior is essential for the widespread acceptance of robots in social environments. Recent advancements in learned legged locomotion controllers have enabled more dynamic and versatile robot behaviors. However, determining the optimal behavior for interactions with different users across varied scenarios remains a challenge. Current methods either rely on natural language inpu… ▽ More Expressive robotic behavior is essential for the widespread acceptance of robots in social environments. Recent advancements in learned legged locomotion controllers have enabled more dynamic and versatile robot behaviors. However, determining the optimal behavior for interactions with different users across varied scenarios remains a challenge. Current methods either rely on natural language input, which is efficient but low-resolution, or learn from human preferences, which, although high-resolution, is sample inefficient. This paper introduces a novel approach that leverages priors generated by pre-trained LLMs alongside the precision of preference learning. Our method, termed Language-Guided Preference Learning (LGPL), uses LLMs to generate initial behavior samples, which are then refined through preference-based feedback to learn behaviors that closely align with human expectations. Our core insight is that LLMs can guide the sampling process for preference learning, leading to a substantial improvement in sample efficiency. We demonstrate that LGPL can quickly learn accurate and expressive behaviors with as few as four queries, outperforming both purely language-parameterized models and traditional preference learning approaches. Website with videos: https://lgpl-gaits.github.io/ △ Less

Submitted 31 March, 2025; v1 submitted 5 February, 2025; originally announced February 2025.

Comments: 8 pages 5 figures

Quest Love: A First Look at Blockchain Loyalty Programs

Authors: Joseph Al-Chami, Jeremy Clark

Abstract: Blockchain ecosystems -- such as those built around chains, layers, and services -- try to engage users for a variety of reasons: user education, growing and protecting their market share, climbing metric-measuring leaderboards with competing systems, demonstrating usage to investors, and identifying worthy recipients for newly created tokens (airdrops). A popular approach is offering user quests:… ▽ More Blockchain ecosystems -- such as those built around chains, layers, and services -- try to engage users for a variety of reasons: user education, growing and protecting their market share, climbing metric-measuring leaderboards with competing systems, demonstrating usage to investors, and identifying worthy recipients for newly created tokens (airdrops). A popular approach is offering user quests: small tasks that can be completed by a user, exposing them to a common task they might want to do in the future, and rewarding them for completion. In this paper, we analyze a proprietary dataset from one deployed quest system that offered 43 unique quests over 10 months with 80M completions. We offer insights about the factors that correlate with task completion: amount of reward, monetary value of reward, difficulty, and cost. We also discuss the role of farming and bots, and the factors that complicate distinguishing real users from automated scripts. △ Less

Submitted 19 March, 2025; v1 submitted 30 January, 2025; originally announced January 2025.

Zero-shot Shark Tracking and Biometrics from Aerial Imagery

Authors: Chinmay K Lalgudi, Mark E Leone, Jaden V Clark, Sergio Madrigal-Mora, Mario Espinoza

Abstract: The recent widespread adoption of drones for studying marine animals provides opportunities for deriving biological information from aerial imagery. The large scale of imagery data acquired from drones is well suited for machine learning (ML) analysis. Development of ML models for analyzing marine animal aerial imagery has followed the classical paradigm of training, testing, and deploying a new m… ▽ More The recent widespread adoption of drones for studying marine animals provides opportunities for deriving biological information from aerial imagery. The large scale of imagery data acquired from drones is well suited for machine learning (ML) analysis. Development of ML models for analyzing marine animal aerial imagery has followed the classical paradigm of training, testing, and deploying a new model for each dataset, requiring significant time, human effort, and ML expertise. We introduce Frame Level ALIgment and tRacking (FLAIR), which leverages the video understanding of Segment Anything Model 2 (SAM2) and the vision-language capabilities of Contrastive Language-Image Pre-training (CLIP). FLAIR takes a drone video as input and outputs segmentation masks of the species of interest across the video. Notably, FLAIR leverages a zero-shot approach, eliminating the need for labeled data, training a new model, or fine-tuning an existing model to generalize to other species. With a dataset of 18,000 drone images of Pacific nurse sharks, we trained state-of-the-art object detection models to compare against FLAIR. We show that FLAIR massively outperforms these object detectors and performs competitively against two human-in-the-loop methods for prompting SAM2, achieving a Dice score of 0.81. FLAIR readily generalizes to other shark species without additional human effort and can be combined with novel heuristics to automatically extract relevant information including length and tailbeat frequency. FLAIR has significant potential to accelerate aerial imagery analysis workflows, requiring markedly less human effort and expertise than traditional machine learning workflows, while achieving superior accuracy. By reducing the effort required for aerial imagery analysis, FLAIR allows scientists to spend more time interpreting results and deriving insights about marine ecosystems. △ Less

Submitted 10 January, 2025; originally announced January 2025.

Clio: Privacy-Preserving Insights into Real-World AI Use

Authors: Alex Tamkin, Miles McCain, Kunal Handa, Esin Durmus, Liane Lovitt, Ankur Rathi, Saffron Huang, Alfred Mountfield, Jerry Hong, Stuart Ritchie, Michael Stern, Brian Clarke, Landon Goldberg, Theodore R. Sumers, Jared Mueller, William McEachen, Wes Mitchell, Shan Carter, Jack Clark, Jared Kaplan, Deep Ganguli

Abstract: How are AI assistants being used in the real world? While model providers in theory have a window into this impact via their users' data, both privacy concerns and practical challenges have made analyzing this data difficult. To address these issues, we present Clio (Claude insights and observations), a privacy-preserving platform that uses AI assistants themselves to analyze and surface aggregate… ▽ More How are AI assistants being used in the real world? While model providers in theory have a window into this impact via their users' data, both privacy concerns and practical challenges have made analyzing this data difficult. To address these issues, we present Clio (Claude insights and observations), a privacy-preserving platform that uses AI assistants themselves to analyze and surface aggregated usage patterns across millions of conversations, without the need for human reviewers to read raw conversations. We validate this can be done with a high degree of accuracy and privacy by conducting extensive evaluations. We demonstrate Clio's usefulness in two broad ways. First, we share insights about how models are being used in the real world from one million Claude.ai Free and Pro conversations, ranging from providing advice on hairstyles to providing guidance on Git operations and concepts. We also identify the most common high-level use cases on Claude.ai (coding, writing, and research tasks) as well as patterns that differ across languages (e.g., conversations in Japanese discuss elder care and aging populations at higher-than-typical rates). Second, we use Clio to make our systems safer by identifying coordinated attempts to abuse our systems, monitoring for unknown unknowns during critical periods like launches of new capabilities or major world events, and improving our existing monitoring systems. We also discuss the limitations of our approach, as well as risks and ethical concerns. By enabling analysis of real-world AI usage, Clio provides a scalable platform for empirically grounded AI safety and governance. △ Less

Submitted 18 December, 2024; originally announced December 2024.

SEMU-Net: A Segmentation-based Corrector for Fabrication Process Variations of Nanophotonics with Microscopic Images

Authors: Rambod Azimi, Yijian Kong, Dusan Gostimirovic, James J. Clark, Odile Liboiron-Ladouceur

Abstract: Integrated silicon photonic devices, which manipulate light to transmit and process information on a silicon-on-insulator chip, are highly sensitive to structural variations. Minor deviations during nanofabrication-the precise process of building structures at the nanometer scale-such as over- or under-etching, corner rounding, and unintended defects, can significantly impact performance. To addre… ▽ More Integrated silicon photonic devices, which manipulate light to transmit and process information on a silicon-on-insulator chip, are highly sensitive to structural variations. Minor deviations during nanofabrication-the precise process of building structures at the nanometer scale-such as over- or under-etching, corner rounding, and unintended defects, can significantly impact performance. To address these challenges, we introduce SEMU-Net, a comprehensive set of methods that automatically segments scanning electron microscope images (SEM) and uses them to train two deep neural network models based on U-Net and its variants. The predictor model anticipates fabrication-induced variations, while the corrector model adjusts the design to address these issues, ensuring that the final fabricated structures closely align with the intended specifications. Experimental results show that the segmentation U-Net reaches an average IoU score of 99.30%, while the corrector attention U-Net in a tandem architecture achieves an average IoU score of 98.67%. △ Less

Submitted 25 November, 2024; originally announced November 2024.

Comments: Accepted to WACV 2025

Decoupling Training-Free Guided Diffusion by ADMM

Authors: Youyuan Zhang, Zehua Liu, Zenan Li, Zhaoyu Li, James J. Clark, Xujie Si

Abstract: In this paper, we consider the conditional generation problem by guiding off-the-shelf unconditional diffusion models with differentiable loss functions in a plug-and-play fashion. While previous research has primarily focused on balancing the unconditional diffusion model and the guided loss through a tuned weight hyperparameter, we propose a novel framework that distinctly decouples these two co… ▽ More In this paper, we consider the conditional generation problem by guiding off-the-shelf unconditional diffusion models with differentiable loss functions in a plug-and-play fashion. While previous research has primarily focused on balancing the unconditional diffusion model and the guided loss through a tuned weight hyperparameter, we propose a novel framework that distinctly decouples these two components. Specifically, we introduce two variables ${x}$ and ${z}$, to represent the generated samples governed by the unconditional generation model and the guidance function, respectively. This decoupling reformulates conditional generation into two manageable subproblems, unified by the constraint ${x} = {z}$. Leveraging this setup, we develop a new algorithm based on the Alternating Direction Method of Multipliers (ADMM) to adaptively balance these components. Additionally, we establish the equivalence between the diffusion reverse step and the proximal operator of ADMM and provide a detailed convergence analysis of our algorithm under certain mild assumptions. Our experiments demonstrate that our proposed method ADMMDiff consistently generates high-quality samples while ensuring strong adherence to the conditioning criteria. It outperforms existing methods across a range of conditional generation tasks, including image generation with various guidance and controllable motion synthesis. △ Less

Submitted 18 November, 2024; originally announced November 2024.

Exploring the Requirements of Clinicians for Explainable AI Decision Support Systems in Intensive Care

Authors: Jeffrey N. Clark, Matthew Wragg, Emily Nielsen, Miquel Perello-Nieto, Nawid Keshtmand, Michael Ambler, Shiv Sharma, Christopher P. Bourdeaux, Amberly Brigden, Raul Santos-Rodriguez

Abstract: There is a growing need to understand how digital systems can support clinical decision-making, particularly as artificial intelligence (AI) models become increasingly complex and less human-interpretable. This complexity raises concerns about trustworthiness, impacting safe and effective adoption of such technologies. Improved understanding of decision-making processes and requirements for explan… ▽ More There is a growing need to understand how digital systems can support clinical decision-making, particularly as artificial intelligence (AI) models become increasingly complex and less human-interpretable. This complexity raises concerns about trustworthiness, impacting safe and effective adoption of such technologies. Improved understanding of decision-making processes and requirements for explanations coming from decision support tools is a vital component in providing effective explainable solutions. This is particularly relevant in the data-intensive, fast-paced environments of intensive care units (ICUs). To explore these issues, group interviews were conducted with seven ICU clinicians, representing various roles and experience levels. Thematic analysis revealed three core themes: (T1) ICU decision-making relies on a wide range of factors, (T2) the complexity of patient state is challenging for shared decision-making, and (T3) requirements and capabilities of AI decision support systems. We include design recommendations from clinical input, providing insights to inform future AI systems for intensive care. △ Less

Submitted 18 November, 2024; originally announced November 2024.

Science Out of Its Ivory Tower: Improving Accessibility with Reinforcement Learning

Authors: Haining Wang, Jason Clark, Hannah McKelvey, Leila Sterman, Zheng Gao, Zuoyu Tian, Sandra Kübler, Xiaozhong Liu

Abstract: A vast amount of scholarly work is published daily, yet much of it remains inaccessible to the general public due to dense jargon and complex language. To address this challenge in science communication, we introduce a reinforcement learning framework that fine-tunes a language model to rewrite scholarly abstracts into more comprehensible versions. Guided by a carefully balanced combination of wor… ▽ More A vast amount of scholarly work is published daily, yet much of it remains inaccessible to the general public due to dense jargon and complex language. To address this challenge in science communication, we introduce a reinforcement learning framework that fine-tunes a language model to rewrite scholarly abstracts into more comprehensible versions. Guided by a carefully balanced combination of word- and sentence-level accessibility rewards, our language model effectively substitutes technical terms with more accessible alternatives, a task which models supervised fine-tuned or guided by conventional readability measures struggle to accomplish. Our best model adjusts the readability level of scholarly abstracts by approximately six U.S. grade levels -- in other words, from a postgraduate to a high school level. This translates to roughly a 90% relative boost over the supervised fine-tuning baseline, all while maintaining factual accuracy and high-quality language. An in-depth analysis of our approach shows that balanced rewards lead to systematic modifications in the base model, likely contributing to smoother optimization and superior performance. We envision this work as a step toward bridging the gap between scholarly research and the general public, particularly younger readers and those without a college degree. △ Less

Submitted 16 April, 2025; v1 submitted 22 October, 2024; originally announced October 2024.

GTQCP: Greedy Topology-Aware Quantum Circuit Partitioning

Authors: Joseph Clark, Travis S. Humble, Himanshu Thapliyal

Abstract: We propose Greedy Topology-Aware Quantum Circuit Partitioning (GTQCP), a novel quantum gate circuit partitioning method which partitions circuits by applying a greedy heuristic to the qubit dependency graph of the circuit. GTQCP is compared against three other gate partitioning methods, two of which (QuickPartitioner and ScanPartitioner) are part of the Berkley Quantum Synthesis Toolkit. GTQCP is… ▽ More We propose Greedy Topology-Aware Quantum Circuit Partitioning (GTQCP), a novel quantum gate circuit partitioning method which partitions circuits by applying a greedy heuristic to the qubit dependency graph of the circuit. GTQCP is compared against three other gate partitioning methods, two of which (QuickPartitioner and ScanPartitioner) are part of the Berkley Quantum Synthesis Toolkit. GTQCP is shown to have 18% run time improvement ratio over the fastest approach (QuickPartitioner), and a 96% improvement over the highest quality approach (ScanPartitioner). The algorithm also demonstrates nearly identical result quality (number of partitions) compared with ScanPartitioner, and a 38% quality improvement over QuickPartitioner. △ Less

Submitted 3 October, 2024; originally announced October 2024.

Comments: 6 pages, 4 figures, 3 tables

Journal ref: 2023 IEEE International Conference on Quantum Computing and Engineering (QCE), 2023, pp. 739-744

Designing Interfaces for Multimodal Vector Search Applications

Authors: Owen Pendrigh Elliott, Tom Hamer, Jesse Clark

Abstract: Multimodal vector search offers a new paradigm for information retrieval by exposing numerous pieces of functionality which are not possible in traditional lexical search engines. While multimodal vector search can be treated as a drop in replacement for these traditional systems, the experience can be significantly enhanced by leveraging the unique capabilities of multimodal search. Central to an… ▽ More Multimodal vector search offers a new paradigm for information retrieval by exposing numerous pieces of functionality which are not possible in traditional lexical search engines. While multimodal vector search can be treated as a drop in replacement for these traditional systems, the experience can be significantly enhanced by leveraging the unique capabilities of multimodal search. Central to any information retrieval system is a user who expresses an information need, traditional user interfaces with a single search bar allow users to interact with lexical search systems effectively however are not necessarily optimal for multimodal vector search. In this paper we explore novel capabilities of multimodal vector search applications utilising CLIP models and present implementations and design patterns which better allow users to express their information needs and effectively interact with these systems in an information retrieval context. △ Less

Submitted 17 September, 2024; originally announced September 2024.

Comments: 12 pages, 8 figures, CIKM 2024 MMSR Workshop

ACM Class: H.5.2; H.1.1; H.1.2; H.3.3

Improving the Reliability of Quantum Circuits by Evolving Heterogeneous Ensembles

Authors: Owain Parry, John Clark, Phil McMinn

Abstract: Quantum computers can perform certain operations exponentially faster than classical computers, but designing quantum circuits is challenging. To that end, researchers used evolutionary algorithms to produce probabilistic quantum circuits that give the correct output more often than not for any input. They can be executed multiple times, with the outputs combined using a classical method (such as… ▽ More Quantum computers can perform certain operations exponentially faster than classical computers, but designing quantum circuits is challenging. To that end, researchers used evolutionary algorithms to produce probabilistic quantum circuits that give the correct output more often than not for any input. They can be executed multiple times, with the outputs combined using a classical method (such as voting) to produce the final output, effectively creating a homogeneous ensemble of circuits (i.e., all identical). Inspired by n-version programming and ensemble learning, we developed a tool that uses an evolutionary algorithm to generate heterogeneous ensembles of circuits (i.e., all different), named QuEEn. We used it to evolve ensembles to solve the Iris classification problem. When using ideal simulation, we found the performance of heterogeneous ensembles to be greater than that of homogeneous ensembles to a statistically significant degree. When using noisy simulation, we still observed a statistically significant improvement in the majority of cases. Our results indicate that evolving heterogeneous ensembles is an effective strategy for improving the reliability of quantum circuits. This is particularly relevant in the current NISQ era of quantum computing where computers do not yet have good tolerance to quantum noise. △ Less

Submitted 13 September, 2024; originally announced September 2024.

Peephole Optimization for Quantum Approximate Synthesis

Authors: Joseph Clark, Himanshu Thapliyal

Abstract: Peephole optimization of quantum circuits provides a method of leveraging standard circuit synthesis approaches into scalable quantum circuit optimization. One application of this technique partitions an entire circuit into a series of peepholes and produces multiple approximations of each partitioned subcircuit. A single approximation of each subcircuit is then selected to form optimized result c… ▽ More Peephole optimization of quantum circuits provides a method of leveraging standard circuit synthesis approaches into scalable quantum circuit optimization. One application of this technique partitions an entire circuit into a series of peepholes and produces multiple approximations of each partitioned subcircuit. A single approximation of each subcircuit is then selected to form optimized result circuits. We propose a series of improvements to the final phase of this architecture, which include the addition of error awareness and a better method of approximating the correctness of the result. We evaluated these proposed improvements on a set of benchmark circuits using the IBMQ FakeWashington simulator. The results demonstrate that our best-performing method provides an average reduction in Total Variational Distance (TVD) and Jensen-Shannon Divergence (JSD) of 18.2% and 15.8%, respectively, compared with the Qiskit optimizer. This also constitutes an improvement in TVD of 11.4% and JSD of 9.0% over existing solutions. △ Less

Submitted 9 September, 2024; originally announced September 2024.

Comments: 8 pages, 4 figures, 1 table

Journal ref: 2024 25th International Symposium on Quality Electronic Design (ISQED), 2024, pp. 1-8

Simplifying Scholarly Abstracts for Accessible Digital Libraries

Authors: Haining Wang, Jason Clark

Abstract: Standing at the forefront of knowledge dissemination, digital libraries curate vast collections of scientific literature. However, these scholarly writings are often laden with jargon and tailored for domain experts rather than the general public. As librarians, we strive to offer services to a diverse audience, including those with lower reading levels. To extend our services beyond mere access,… ▽ More Standing at the forefront of knowledge dissemination, digital libraries curate vast collections of scientific literature. However, these scholarly writings are often laden with jargon and tailored for domain experts rather than the general public. As librarians, we strive to offer services to a diverse audience, including those with lower reading levels. To extend our services beyond mere access, we propose fine-tuning a language model to rewrite scholarly abstracts into more comprehensible versions, thereby making scholarly literature more accessible when requested. We began by introducing a corpus specifically designed for training models to simplify scholarly abstracts. This corpus consists of over three thousand pairs of abstracts and significance statements from diverse disciplines. We then fine-tuned four language models using this corpus. The outputs from the models were subsequently examined both quantitatively for accessibility and semantic coherence, and qualitatively for language quality, faithfulness, and completeness. Our findings show that the resulting models can improve readability by over three grade levels, while maintaining fidelity to the original content. Although commercial state-of-the-art models still hold an edge, our models are much more compact, can be deployed locally in an affordable manner, and alleviate the privacy concerns associated with using commercial models. We envision this work as a step toward more inclusive and accessible libraries, improving our services for young readers and those without a college degree. △ Less

Submitted 7 August, 2024; originally announced August 2024.

Comments: Initial submission to JCDL2024

Towards Personalised Patient Risk Prediction Using Temporal Hospital Data Trajectories

Authors: Thea Barnes, Enrico Werner, Jeffrey N. Clark, Raul Santos-Rodriguez

Abstract: Quantifying a patient's health status provides clinicians with insight into patient risk, and the ability to better triage and manage resources. Early Warning Scores (EWS) are widely deployed to measure overall health status, and risk of adverse outcomes, in hospital patients. However, current EWS are limited both by their lack of personalisation and use of static observations. We propose a pipeli… ▽ More Quantifying a patient's health status provides clinicians with insight into patient risk, and the ability to better triage and manage resources. Early Warning Scores (EWS) are widely deployed to measure overall health status, and risk of adverse outcomes, in hospital patients. However, current EWS are limited both by their lack of personalisation and use of static observations. We propose a pipeline that groups intensive care unit patients by the trajectories of observations data throughout their stay as a basis for the development of personalised risk predictions. Feature importance is considered to provide model explainability. Using the MIMIC-IV dataset, six clusters were identified, capturing differences in disease codes, observations, lengths of admissions and outcomes. Applying the pipeline to data from just the first four hours of each ICU stay assigns the majority of patients to the same cluster as when the entire stay duration is considered. In-hospital mortality prediction models trained on individual clusters had higher F1 score performance in five of the six clusters when compared against the unclustered patient cohort. The pipeline could form the basis of a clinical decision support tool, working to improve the clinical characterisation of risk groups and the early detection of patient deterioration. △ Less

Submitted 12 July, 2024; originally announced July 2024.

Automatic Pruning of Fine-tuning Datasets for Transformer-based Language Models

Authors: Mohammadreza Tayaranian, Seyyed Hasan Mozafari, Brett H. Meyer, James J. Clark, Warren J. Gross

Abstract: Transformer-based language models have shown state-of-the-art performance on a variety of natural language understanding tasks. To achieve this performance, these models are first pre-trained on general corpus and then fine-tuned on downstream tasks. Previous work studied the effect of pruning the training set of the downstream tasks on the performance of the model on its evaluation set. In this w… ▽ More Transformer-based language models have shown state-of-the-art performance on a variety of natural language understanding tasks. To achieve this performance, these models are first pre-trained on general corpus and then fine-tuned on downstream tasks. Previous work studied the effect of pruning the training set of the downstream tasks on the performance of the model on its evaluation set. In this work, we propose an automatic dataset pruning method for the training set of fine-tuning tasks. Our method is based on the model's success rate in correctly classifying each training data point. Unlike previous work which relies on user feedback to determine subset size, our method automatically extracts training subsets that are adapted for each pair of model and fine-tuning task. Our method provides multiple subsets for use in dataset pruning that navigate the trade-off between subset size and evaluation accuracy. Our largest subset, which we also refer to as the winning ticket subset, is on average $3 \times$ smaller than the original training set of the fine-tuning task. Our experiments on 5 downstream tasks and 2 language models show that, on average, fine-tuning on the winning ticket subsets results in a $0.1 \%$ increase in the evaluation performance of the model. △ Less

Submitted 11 July, 2024; originally announced July 2024.

Comments: 28 pages, 17 figures. Accepted at the Third Conference on Lifelong Learning Agents (CoLLAs 2024)

Artificial Intelligence Index Report 2024

Authors: Nestor Maslej, Loredana Fattorini, Raymond Perrault, Vanessa Parli, Anka Reuel, Erik Brynjolfsson, John Etchemendy, Katrina Ligett, Terah Lyons, James Manyika, Juan Carlos Niebles, Yoav Shoham, Russell Wald, Jack Clark

Abstract: The 2024 Index is our most comprehensive to date and arrives at an important moment when AI's influence on society has never been more pronounced. This year, we have broadened our scope to more extensively cover essential trends such as technical advancements in AI, public perceptions of the technology, and the geopolitical dynamics surrounding its development. Featuring more original data than ev… ▽ More The 2024 Index is our most comprehensive to date and arrives at an important moment when AI's influence on society has never been more pronounced. This year, we have broadened our scope to more extensively cover essential trends such as technical advancements in AI, public perceptions of the technology, and the geopolitical dynamics surrounding its development. Featuring more original data than ever before, this edition introduces new estimates on AI training costs, detailed analyses of the responsible AI landscape, and an entirely new chapter dedicated to AI's impact on science and medicine. The AI Index report tracks, collates, distills, and visualizes data related to artificial intelligence (AI). Our mission is to provide unbiased, rigorously vetted, broadly sourced data in order for policymakers, researchers, executives, journalists, and the general public to develop a more thorough and nuanced understanding of the complex field of AI. The AI Index is recognized globally as one of the most credible and authoritative sources for data and insights on artificial intelligence. Previous editions have been cited in major newspapers, including the The New York Times, Bloomberg, and The Guardian, have amassed hundreds of academic citations, and been referenced by high-level policymakers in the United States, the United Kingdom, and the European Union, among other places. This year's edition surpasses all previous ones in size, scale, and scope, reflecting the growing significance that AI is coming to hold in all of our lives. △ Less

Submitted 29 May, 2024; originally announced May 2024.

The Impacts of Data, Ordering, and Intrinsic Dimensionality on Recall in Hierarchical Navigable Small Worlds

Authors: Owen Pendrigh Elliott, Jesse Clark

Abstract: Vector search systems, pivotal in AI applications, often rely on the Hierarchical Navigable Small Worlds (HNSW) algorithm. However, the behaviour of HNSW under real-world scenarios using vectors generated with deep learning models remains under-explored. Existing Approximate Nearest Neighbours (ANN) benchmarks and research typically has an over-reliance on simplistic datasets like MNIST or SIFT1M… ▽ More Vector search systems, pivotal in AI applications, often rely on the Hierarchical Navigable Small Worlds (HNSW) algorithm. However, the behaviour of HNSW under real-world scenarios using vectors generated with deep learning models remains under-explored. Existing Approximate Nearest Neighbours (ANN) benchmarks and research typically has an over-reliance on simplistic datasets like MNIST or SIFT1M and fail to reflect the complexity of current use-cases. Our investigation focuses on HNSW's efficacy across a spectrum of datasets, including synthetic vectors tailored to mimic specific intrinsic dimensionalities, widely-used retrieval benchmarks with popular embedding models, and proprietary e-commerce image data with CLIP models. We survey the most popular HNSW vector databases and collate their default parameters to provide a realistic fixed parameterisation for the duration of the paper. We discover that the recall of approximate HNSW search, in comparison to exact K Nearest Neighbours (KNN) search, is linked to the vector space's intrinsic dimensionality and significantly influenced by the data insertion sequence. Our methodology highlights how insertion order, informed by measurable properties such as the pointwise Local Intrinsic Dimensionality (LID) or known categories, can shift recall by up to 12 percentage points. We also observe that running popular benchmark datasets with HNSW instead of KNN can shift rankings by up to three positions for some models. This work underscores the need for more nuanced benchmarks and design considerations in developing robust vector search systems using approximate vector search algorithms. This study presents a number of scenarios with varying real world applicability which aim to better increase understanding and future development of ANN algorithms and embedding △ Less

Submitted 28 May, 2024; originally announced May 2024.

Comments: 15 pages, 2 figures

Design Editing for Offline Model-based Optimization

Authors: Ye Yuan, Youyuan Zhang, Can Chen, Haolun Wu, Zixuan Li, Jianmo Li, James J. Clark, Xue Liu

Abstract: Offline model-based optimization (MBO) aims to maximize a black-box objective function using only an offline dataset of designs and scores. These tasks span various domains, such as robotics, material design, and protein and molecular engineering. A common approach involves training a surrogate model using existing designs and their corresponding scores, and then generating new designs through gra… ▽ More Offline model-based optimization (MBO) aims to maximize a black-box objective function using only an offline dataset of designs and scores. These tasks span various domains, such as robotics, material design, and protein and molecular engineering. A common approach involves training a surrogate model using existing designs and their corresponding scores, and then generating new designs through gradient-based updates with respect to the surrogate model. This method suffers from the out-of-distribution issue, where the surrogate model may erroneously predict high scores for unseen designs. To address this challenge, we introduce a novel method, Design Editing for Offline Model-based Optimization (DEMO), which leverages a diffusion prior to calibrate overly optimized designs. DEMO first generates pseudo design candidates by performing gradient ascent with respect to a surrogate model. While these pseudo design candidates contain information beyond the offline dataset, they might be invalid or have erroneously high predicted scores. Therefore, to address this challenge while utilizing the information provided by pseudo design candidates, we propose an editing process to refine these pseudo design candidates. We introduce noise to the pseudo design candidates and subsequently denoise them with a diffusion prior trained on the offline dataset, ensuring they align with the distribution of valid designs. Empirical evaluations on seven offline MBO tasks show that, with properly tuned hyperparameters, DEMOs score is competitive with the best previously reported scores in the literature. △ Less

Submitted 17 April, 2025; v1 submitted 22 May, 2024; originally announced May 2024.

Comments: Accepted by Transactions on Machine Learning Research (TMLR)