Fast Machine Learning for Science Conference 2026

Session

Reception & Poster Session

31 Aug 2026, 17:36

92. A Binary-Weight Transformer for Real-Time Particle Jet Tagging at the LHC

Kai Yamaguchi

31/08/2026, 17:36

Poster

The LHC collides protons at a rate of 40 million collisions per second. To filter the massive amount of data for interesting physics, the real-time trigger systems inside detectors at the LHC necessitate smart and sophisticated triggers that are 1) efficient enough to simultaneously reject large backgrounds and keep enough signal, 2) compact enough to meet hardware constraints, and 3) fast...

93. A Controlled Benchmark for Memory Mechanisms in Efficient-Attention Transformers

Robert Sneiderman (Independent)

31/08/2026, 17:36

Poster

Efficient-attention transformers trade direct context access for compute, and a family of memory mechanisms is offered to recover what restricted attention drops. Reported gains are hard to trust: papers vary the mechanism, the backbone, the optimizer, the token budget, and the scale all at once, and they report a single aggregate perplexity that averages over positions where local attention...

69. A Controlled Comparison of FINN and Vitis AI for Small Neural Networks

Ruthwik Reddy Sunketa

31/08/2026, 17:36

Poster

Small neural networks are an important class of FPGA workload. Many scientific and edge applications run small models under tight latency budgets, including high-energy-physics triggering, network-intrusion detection, and keyword spotting. When the model is small enough to keep on-chip, FPGAs can deliver low-latency, real-time inference. Two widely-used toolchains approach deployment...

43. A Mamba-2 based Selective State Space Model for Improving MET Estimation in CMS Run 3 and the HL-LHC

Aleesha Kallil Tharayil (Carnegie-Mellon University (US))

31/08/2026, 17:36

Poster

Accurate reconstruction of the missing transverse momentum (MET) is essential for a broad range of CMS analyses, including searches for new physics and precision measurements of the W boson mass. However, its resolution degrades significantly in high-pileup environments and is expected to degrade further at the High-Luminosity LHC. A state-of-the-art deep neural network-based MET...

109. Accelerating Lattice Gauge Simulations with Gauge-Equivariant Graph Neural Networks

Gia-Wei Chern (University of Virginia)

31/08/2026, 17:36

Poster

Machine learning is rapidly transforming computational science by replacing expensive first-principles calculations with accurate, scalable surrogate models. A major challenge, however, is incorporating the fundamental physical symmetries that govern scientific simulations. We present a gauge-equivariant graph neural network (GNN) that enables scalable machine learning for lattice gauge...

10. An extreme edge ML-based electron/photon trigger for the ATLAS Global Trigger

Adrian Chitan (IFIN-HH (RO))

31/08/2026, 17:36

Poster

Final states containing isolated electrons and photons (e/γ) have played a central
role in physics discoveries at the Large Hadron Collider (LHC) and will remain
vital to the ATLAS trigger strategy throughout the High-Luminosity LHC (HL-LHC)
programme. However, the substantially increased pile-up expected at the HL-LHC
will make it increasingly challenging to preserve high trigger...

18. Applications of machine learning to classification of cosmic X-ray sources: results and challenges.

Yichao Lin (The George Washington University)

31/08/2026, 17:36

Poster

Over the past 25 years, over 2 million X-ray sources have been serendipitously discovered by various X-ray observatories, where the majority remain unclassified. Traditional manual classification methods alone are increasingly unable to keep up with the growth of data. We present the results and lessons learned from applying a random forest classifier to X-ray sources from the XMM-Newton...

45. Benchmarking BitNet-Style Low-Precision Networks for Fast Jet Classification Toward FPGA Deployment

Dorian Sloot (Austrian Academy of Sciences (AT))

31/08/2026, 17:36

Poster

Low-latency machine learning is essential for real-time decision systems in high-energy physics, where strict resource and latency constraints limit deployable model complexity. Aggressive quantization can potentially reduce memory and arithmetic requirements, but its effect on predictive performance must be understood before deployment. We present a controlled comparison of low-precision...

83. Bridging the Gap Between Enterprise Memory-Intensive Workloads and Real-Time Distributed ML Systems: A Scheduling Reality Check

Atul Garg

31/08/2026, 17:36

Poster

Distributed Machine Learning (ML) systems and real-time inference pipelines are traditionally designed and benchmarked under the assumption of homogeneous, dedicated cloud infrastructure. However, in modern corporate and sovereign cloud environments, distributed ML workloads must routinely co-exist alongside long-running, memory-heavy enterprise resource planning (ERP) databases and...

49. CNN-based event classification for first level triggers at HL-LHC

Jim Brooke (University of Bristol (GB))

31/08/2026, 17:36

Poster

We describe a CNN based approach to full event classification, for fast first level event selection at the HL-LHC. We perform hardware-aware optimisation of the network architecture, and evaluate physics performance using simulated data. This allowed a range of network models to be identified that fit within target FPGA resources and latency requirements of HL-LHC trigger systems. A candidate...

36. Deploying a Diffusion model on Voyager for cosmological applications

Javier Hernandez-Nicolau (San Diego Supercomputer Center)

31/08/2026, 17:36

Poster

The integration of AI into cosmological research is poised to significantly impact major experiments such as Simons Observatory, LiteBIRD, and CMB-S4. These projects aim to achieve unprecedented precision in mapping the cosmic microwave background (CMB), necessitating high-resolution simulations to interpret the data accurately. AI techniques, particularly those enhancing low-resolution...

71. Design and Implementation of Tiny Real-Time ML for FPGAs with Agentic Workflows

Qibin Liu (SLAC National Accelerator Laboratory (US))

31/08/2026, 17:36

Poster

Recent advances in machine learning and microelectronics are enabling efficient real-time, on-chip data processing under stringent latency, power, and bandwidth constraints. Compact machine-learning models implemented directly in hardware can replace or augment fixed logic for intelligent feature extraction, classification, and denoising at the detector front-end. These capabilities are...

60. Direct Parameter Estimation of Binary Neutron Star Mergers with State Space Models

Kyungseop Yoon (Massachusetts Institute of Technology)

31/08/2026, 17:36

Poster

Fast and accurate parameter estimation of binary neutron star (BNS) mergers, gravitational wave events with electromagnetic counterparts, remains a central challenge in multimessenger astronomy. Building on state space models (SSMs), we directly regress BNS merger source parameters from raw gravitational wave time series, without sampling-based inference. As a first demonstration, we focus on...

80. ErwinParT: A Radially-Partitioned Hierarchical Transformer for Ultra-Low Latency Jet Tagging at the CMS Level-1 Trigger

Vaibhav Lohia

31/08/2026, 17:36

Poster

Real-time identification of jets from boosted heavy particles is vital for the High-Luminosity LHC physics program at CMS. However, deploying highly expressive architectures like the Particle Transformer within the hardware-constrained Level-1 (L1) trigger is severely limited by the quadratic scaling $\mathcal{O}(N^2)$ of standard self-attention. This computational burden makes...

37. Fast machine learning trigger system for real-time event identification for Ba-tagging in a Liquid Xenon TPC

Melissa Medina Peregrina

31/08/2026, 17:36

Poster

A next-generation neutrinoless double beta decay (0νββ) search in ¹³⁶Xe has the potential to uncover lepton number violation, and determine if neutrinos are their own antiparticle. This rare decay, if discovered, would demonstrate Physics beyond the Standard Model and provide key insights into the evolution of the universe.

Fast machine learning allows real time data processing in hardware,...

41. FlowFI: Dataflow and Quantization Aware Fault Injection for Resilient Accelerators

Raymond Duenas

31/08/2026, 17:36

Poster

In high radiation environments, hardware accelerators are prone to radiation-induced bit flips, leading to data corruption. For example, scientists at the Large Hadron Collider (LHC) seek to deploy hardware-accelerated neural networks in environments with radiation 1000X higher than that seen in space [1]. Ensuring reliable data collection, such as at the LHC, requires developing hardware that...

54. FQTree: Fine-Grained Quantization and Hardware Generation of Boosted Decision Trees

Zhiqiang Que (University of Bristol), Chang Sun (California Institute of Technology (US))

31/08/2026, 17:36

Poster

Boosted decision trees (BDTs) are widely used in latency-critical applications, but efficient FPGA deployment remains challenging. Existing designs often rely on uniform or manually tuned fixed-point formats, which can introduce unnecessary hardware cost or accuracy loss. This work presents FQTree, a framework combining fine-grained quantization-aware training with automatic hardware...

28. From Keras 2 to Keras 3: QKerasV3 Integration with the hls4ml FPGA Flow

Marius Köppel (ETH Zurich (CH))

31/08/2026, 17:36

Poster

Quantization‑aware training (QAT) is essential for delivering low‑latency inference on the FPGAs that power high‑energy‑physics (HEP) experiments. QKeras has become the de‑facto Keras‑based framework for QAT in this community and is tightly coupled with the hls4ml toolchain, which converts Keras models into synthesizable HLS code. The transition from Keras 2 to Keras 3 introduced a new "ops"...

105. Generative Cosmological Emulators as Fluid Flows in Data Space

Sandip Roy (University of California San Diego)

31/08/2026, 17:36

Poster

Cosmological simulations are too computationally expensive to support exhaustive scans over dark matter parameter space. Generative models, including diffusion models and flow models, offer a promising route to simulation emulation using existing suites of numerical simulations. In this talk, I will describe diffusion and flow-based generative models as transport processes: they evolve...

77. HACE: A Surrogate-Guided Agent for hls4ml Configuration on Scientific FPGAs

Mr Venkata Sai Prathyush Turaga (Texas Tech University)

31/08/2026, 17:36

Poster

Configuring hls4ml(High-Level Synthesis for Machine Learning) for a new scientific task is a manual loop. The user sets precision, reuse factor, strategy, IO type, and backend, runs HLS synthesis, and reads back latency and resource usage. A single synthesis pass takes minutes to hours, and the search space grows quickly when settings are tuned per layer. Domain experts in particle physics,...

42. Jet Tagger for Real-Time Triggering at the LHC

Kalib McEuen (Univ. of Illinois Chicago (US))

31/08/2026, 17:36

Poster

The Large Hadron Collider (LHC) collides protons at a rate of 40 million collisions per second. To filter the massive amount of data for interesting physics, the real-time trigger systems inside detectors at the LHC necessitate smart and sophisticated triggers that are 1) efficient enough to simultaneously reject large backgrounds and keep enough signal, 2) compact enough to meet hardware...

102. Lightweight Transformers for High-Energy Physics Experiments

Erdem Yigit Ertorer (Carnegie-Mellon University (US))

31/08/2026, 17:36

Poster

Transformers have gained significant traction in high-energy physics (HEP) experiments. However, many HEP applications require ultra-fast processing and the quadratic complexity of standard self-attention becomes prohibitive. Fortunately, there are several approaches to address this bottleneck. One is to replace this standard self-attention with linearized attention. In this work, we explore...

38. Machine Learning-Based Large-R Jet Tagging for the ATLAS Level-0 Global Trigger

Maryam Bayat Makou (Southern Methodist University (US))

31/08/2026, 17:36

Poster

Efficient identification of boosted hadronic objects is an important challenge for the ATLAS Phase-II trigger system. This work investigates machine-learning-based large-R jet tagging for the ATLAS Level-0 Global Trigger using trigger-level calorimeter information under HL-LHC pile-up conditions.
Using simulated $HH\rightarrow b\bar{b}b\bar{b}$ signal and QCD multijet background events, we...

25. Machine-Learning-Based Self-Trigger for the COHERENT Cryogenic CsI Detector

Zepeng Li (University of Hawaii at Manoa)

31/08/2026, 17:36

Poster

The COHERENT experiment has demonstrated coherent elastic neutrino–nucleus scattering (CEvNS) at the Spallation Neutron Source, establishing a powerful neutral-current channel for probing all neutrino flavors. Tonne-scale CEvNS detectors, especially the cryogenic CsI detector, offer a promising opportunity to detect neutrinos from a Galactic core-collapse supernova. A key challenge is that the...

99. Memory-Efficient Activation Checkpointing with Sliding Window and Hirschberg's Algorithm for 0/1 Knapsack Solving in PyTorch

Jędrzej Maczan

31/08/2026, 17:36

Poster

Activation checkpointing minimizes the runtime of neural networks under a given memory budget, by selecting which intermediate tensors to store and which to recompute. PyTorch solves this as a 0/1 knapsack problem, where operations from a joint forward-backward computation graph are items with a memory cost (weight) and a runtime saving (value). The default solver, dp_knapsack, allocates a...

94. ML Classifiers for the search of new post-AGB candidates in all-sky surveys.

Diego Osvaldo Ochoa de la Cruz

31/08/2026, 17:36

Poster

Post-Asymptotic Giant Branch (post-AGB) stars are critical, short-lived transition objects in stellar evolution. However, only 394 confirmed post-AGB stars are currently known, heavily limiting our ability to constrain AGB models. While recent massive all-sky surveys (e.g., 2MASS, WISE, SDSS, IGAPS, and VPHAS+) offer an unprecedented opportunity to discover new candidates at scale, exploiting...

91. Multi-SG Parameter Estimation for Anomalous Gravitational Wave Signals

Jack Redepenning

31/08/2026, 17:36

Poster

Detection of gravitational waves (GWs) has opened new roads in exploring and analyzing astrophysical data. Not only can we learn more about gravitational waves themselves, but this also allows us to perform multi-messenger astronomy, detecting both the GW and electromagnetic (EM) signals. GW170817 demonstrated the power of multi-messenger detections. It confirmed that neutron star mergers...

59. Nereid: an Inference Server for HPC

Ethan Colbert (Purdue University (US)), Noah Paladino (Massachusetts Inst. of Technology (US))

31/08/2026, 17:36

Poster

As scientific data analysis workflows embrace machine learning, demand for elastic GPU compute is rising. Collaborations have begun to turn to inference-as-a-service solutions, which rely on inference servers like NVIDIA Triton, to meet these needs. However, inference servers designed for the cloud have notable shortcomings when deployed within an HPC environment, including difficulties...

33. PermuGini-RF: A Hybrid Feature Selection Model of Permutation and Gini Importance for Mobile Phone Price Classification

FEMI JOHNSON

31/08/2026, 17:36

Poster

The accuracy of machine learning models strongly relies on data quality, including carefully selected features based on established metrics such as Feature importance scores. This paper introduces PermuGini-RF, a hybrid feature selection and classification model that systematically combines Gini Importance for fast upstream screening and Permutation Importance for robust downstream validation...

23. PLANT: Conditional Generative Models for Fast Gravitational-Wave Population Synthesis

Rhea Senthil Kumar (University of California, San Diego)

31/08/2026, 17:36

Poster

Gravitational-wave (GW) observations provide a unique probe of the underlying massive-star population, but extracting this information requires modeling how massive stars evolve into merging compact binaries across cosmic time. Forward population-synthesis frameworks that couple binary evolution to cosmic star-formation and metallicity histories are therefore needed to connect observed merger...

7. Predicting grain-size-aware flow behavior via a physics-informed neural network elasto-viscoplastic model

Adnan Eghtesad

31/08/2026, 17:36

Poster

We introduce a physics-informed elasto-viscoplastic (NN-EVP) framework that utilizes Input Convex Neural Networks (ICNNs) to ensure thermodynamic consistency while maintaining high predictive expressivity. Developed within the PyTorch ecosystem, this automated constitutive modeling tool is validated against both synthetic power-law data and experimental uniaxial deformation data under large...

56. Proxy-guided DSE for Neural Networks on FPGAs

Gabriele Trotta

31/08/2026, 17:36

Poster

As artificial intelligence becomes more capable, it becomes ever more widely adopted, and so do their energy demands. Custom hardware such as FPGAs offers a way to absorb this cost, unveiling a rich design space across which a neural network can be tuned for competing objectives like accuracy, trustworthiness and power. Exploring that space comes with a caveat, however: the processing needed...

87. Pruning LUT-Based Neural Network Ensembles for Efficient FPGA Inference

Sijan Shrestha (Howard University)

31/08/2026, 17:36

Poster

LUT-based Neural Networks (NNs) demonstrate significant potential for low-latency and high-throughput inference on FPGAs in the fields like high-energy physics, high-frequency trading, etc [1][2]. Ensemble approaches like AmigoLUT [3] improve scalability and accuracy of LUT-based NNs such as LogicNets [4] and NeuraLUT [5]. However, increasing ensemble size, even when we start with a small base...

101. Real-Time Multi-Messenger Trigger Fusion Using Machine Learning for Sub-Threshold Event Detection

Mr Kaamesh Chandrasekaran (Sri Venkateswara College of Engineering), Ms Madhushree Naga (Sri Venkateswara College of Engineering)

31/08/2026, 17:36

Poster

Multi-messenger astronomy relies on three independent alert systems - gravitational wave detectors (LIGO/Virgo/KAGRA), high-energy neutrino observatories (IceCube), and gamma-ray monitors (Fermi-GBM), each firing alerts only when their own threshold is individually crossed. The problem is that when all three show near-threshold activity around the same time, no existing pipeline recognizes it...

103. Reject or Compress. Event streaming and triggering system using Byte Latent Transformer.

Leonid Didukh

31/08/2026, 17:36

Poster

Future high-energy physics and gravitational-wave experiments are projected to generate data at unprecedented event rates, demanding fast, scalable, and efficient data management systems. Because scientific data volumes continue to outpace available storage infrastructure, experimental workflows increasingly rely on a combination of real-time triggering mechanisms to filter uninformative...

34. Residual Geometry as a Diagnostic for Hidden Failure Modes in Fast Language Model Compression

Jun-Sik Yoo

31/08/2026, 17:36

Poster

Low-bit quantization is usually judged by whether the compressed model still works: perplexity, output KL, downstream accuracy, or layer reconstruction error. These are useful metrics, but they are also coarse views of a high-dimensional computation. A model can look healthy under these summaries while some part of its internal layer update has changed in a more structured way.

We study...

82. SNAC-Pack: Surrogate Neural Architecture Codesign Package for Hardware-Aware Machine Learning

Jason Weitz (Univ. of California San Diego (US))

31/08/2026, 17:36

Poster

Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and...

15. SparsePixels++: Scalable Sparse Convolution on FPGAs

Ho-Fung Tsoi (University of Pennsylvania)

31/08/2026, 17:36

Poster

Modern particle physics experiments often impose strict latency constraints (microseconds or below) on the edge electronics to extract quality signals from noisy raw data in real time. For 2D image data where signals are spatially sparse, standard CNNs are inefficient because latency and resources scale directly with image size, as every input pixel is densely convolved, including the vast...

96. Sub-100ns On-Probe Spike Sorting via Resource-Constrained FPGA

Aarav Gaur

31/08/2026, 17:36

Poster

On-probe spike sorting aims to discriminate between neurons using electrical signals recorded directly at the probe, enabling real-time applications such as brain-computer interfaces. To generate ground-truth training data, we follow SpikeForest/MEArec-style simulation methodology to construct a synthetic multi-neuron tetrode recording with physiologically-motivated waveforms, controlled SNR,...

104. SUNet: A Spectral UNet for Axion Dark Matter Signal Denoising

Akbota Assan (University of California, San Diego)

31/08/2026, 17:36

Poster

Axion dark matter searches such as ABRACADABRA produce continuous high-
rate time series in which injected signals occupy a single narrow frequency bin
per time frame, a structure that is natural in the spectral domain but opaque in
the raw time domain. Existing TIDMAD denoising approaches either operate
directly on raw time series or require separate model weights per frequency...

110. TensorFlow-to-ONNX Migration of CMS DeepTau Models for Efficient Inference in CMSSW and SONIC

Pritam Palit (Carnegie-Mellon University (US))

31/08/2026, 17:36

Poster

DeepTau is the convolutional neural network (CNN)-based multiclass classifier for hadronic tau identification in CMS. To improve inference performance and simplify deployment, the DeepTau models have been migrated from TensorFlow to ONNX (Open Neural Network Exchange) within the CMS software framework (CMSSW) using ONNX Runtime. In parallel, the TensorFlow-based deployment in SONIC (Services...

44. The CMS ML-based Data Quality Monitoring Infrastructure

Dr Antonio Vagnerini (University of Nebraska-Lincoln)

31/08/2026, 17:36

Poster

The increasing complexity and data throughput of the CMS experiment at the LHC demand scalable and intelligent tools to ensure data quality. In this talk, we present a machine learning-oriented infrastructure designed to support the offline data quality monitoring (DQM) process at CMS. The infrastructure enables the integration of ML algorithms into the DQM workflow, providing auto- mated...

4. To Chop or Not To Chop: the Question for Variational Autoencoders as Anomaly Detectors in High Energy Physics

Sterre Hoogendoorn (University of Pennsylvania)

31/08/2026, 17:36

Poster

Anomaly detection (AD) has recently emerged as an exciting alternative to conventional search strategies in high energy physics. The integration of these techniques into trigger systems is even more recent, but represents a crucial step in expanding the coverage of LHC triggers. In this paper, we explore the direct comparison, as well as combination, of two compression techniques for...

47. TriggerFlow: A Reproducible End-to-End Framework for Machine Learning–Driven Level-1 Trigger Systems in CMS

Marius Köppel (ETH Zurich (CH))

31/08/2026, 17:36

Poster

The CMS Phase-2 upgrade integrates machine learning (ML) throughout the Level-1 Trigger, marking a transition toward differentiable detector systems in which multiple ML models collaboratively reconstruct physics objects in real time. As detector conditions and physics goals evolve, these models require continuous retraining, validation, and deployment, transforming trigger algorithms into...

78. TSCC as an AI-for-Science Platform: Enabling Research with LLMs, Agentic Tools, and Campus Cyberinfrastructure

Subhashini Sivagnanam (San Diego Supercomputer Center), Yuwu Chen (San Diego Supercomputer Center)

31/08/2026, 17:36

Poster

The Triton Shared Computing Cluster (TSCC) at the San Diego Supercomputer Center is evolving beyond a traditional high-performance computing system into an AI-for-science platform that enables researchers to integrate artificial intelligence into scientific discovery. TSCC provides campus researchers with access to computational resources through both Condo (system purchase) and Hotel...

55. Using Variational Autoencoders to Generate Atmospheric Models for Brown Dwarfs

Julia Haynes

31/08/2026, 17:36

Poster

The growing volume of substellar spectra from JWST, including NIRSpec observations of brown dwarfs and directly imaged exoplanets, demands increasingly efficient tools for atmospheric characterization. Traditional spectral fitting approaches such as grid interpolation and Markov Chain Monte Carlo (MCMC) retrieval become significant computational bottlenecks when applied to large samples or...