Machine Learning · Cross-listed from eess.SP
ME-WARD: A multimodal ergonomic analysis tool for musculoskeletal risk assessment from inertial and video data in working plac
Javier González-Alonso, Paula Martín-Tapia, David González-Ortega et al. (+3 more)
This study presents ME-WARD (
Multimodal Ergonomic Workplace Assessment and Risk from Data), a novel system for ergonomic assessment and musculoskeletal risk evaluation that implements the Rapid Upper Limb Assessment (RULA) method. ME-WARD is designed to process joint angle data from motion capture systems, including inertial measurement unit (IMU)-based setups, and deep learning human body pose tracking models. The tool's flexibility enables ergonomic risk assessment using any system capable of reliably measuring joint angles, extending the applicability of RULA beyond proprietary setups. To validate its performance, the tool was tested in an industrial setting during the assembly of conveyor belts, which involved high-risk tasks such as inserting rods and pushing conveyor belt components. The experiments leveraged gold standard IMU systems alongside a state-of-the-art
monocular 3D pose estimation system. The results confirmed that ME-WARD produces reliable RULA scores that closely align with IMU-derived metrics for flexion-dominated movements and comparable performance with the
monocular system, despite limitations in tracking lateral and rotational motions. This work highlights the potential of integrating multiple motion capture technologies into a unified and accessible ergonomic assessment pipeline. By supporting diverse input sources, including low-cost
video-based systems, the proposed
multimodal approach offers a scalable, cost-effective solution for ergonomic assessments, paving the way for broader adoption in resource-constrained industrial environments.
Read the full abstract ▾
All authors Javier González-Alonso, Paula Martín-Tapia, David González-Ortega, Míriam Antón-Rodríguez, Francisco Javier Díaz-Pernas, Mario Martínez-Zarzuela
Machine Learning
Quantum-Inspired Episode Selection for Monte Carlo Reinforcement Learning via QUBO Optimization
Hadi Salloum, Ali Jnadi, Yaroslav Kholodov et al. (+1 more)
Monte Carlo (MC) reinforcement learning suffers from high sample complexity, especially in environments with sparse rewards, large state spaces, and correlated trajectories. We address these limitations by reformulating episode selection as a Quadratic Unconstrained Binary Optimization (QUBO) problem and solving it with
quantum-inspired samplers. Our method, MC+QUBO, integrates a combinatorial filtering step into standard MC policy evaluation: from each batch of trajectories, we select a subset that maximizes cumulative reward while promoting state-space coverage. This selection is encoded as a QUBO, where linear terms favor high-reward episodes and quadratic terms penalize redundancy. We explore both Simulated
Quantum Annealing (SQA) and Simulated Bifurcation (SB) as black-box solvers within this framework. Experiments in a finite-horizon GridWorld demonstrate that MC+QUBO outperforms vanilla MC in convergence speed and final policy quality, highlighting the potential of
quantum-inspired optimization as a decision-making subroutine in reinforcement learning.
Read the full abstract ▾
All authors Hadi Salloum, Ali Jnadi, Yaroslav Kholodov, Alexander Gasnikov
Machine Learning · Cross-listed from cs.AI
JaxARC: A High-Performance JAX-based Environment for Abstraction and Reasoning Research
Aadam, Monu Verma and Mohamed Abdel-Mottaleb
The Abstraction and
Reasoning Corpus (ARC) tests AI systems' ability to perform human-like inductive
reasoning from a few demonstration pairs. Existing Gymnasium-based RL environments severely limit experimental scale due to computational bottlenecks. We present JaxARC, an open-source, high-performance RL environment for ARC implemented in JAX. Its functional, stateless architecture enables massive parallelism, achieving 38-5,439x speedup over Gymnasium at matched batch sizes, with peak throughput of 790M steps/second. JaxARC supports multiple ARC
datasets, flexible action spaces, composable wrappers, and configuration-driven reproducibility, enabling large-scale RL research previously computationally infeasible. JaxARC is available at https://github.com/aadimator/JaxARC.
Read the full abstract ▾
Machine Learning
Towards Generalisable Imitation Learning Through Conditioned Transition Estimation and Online Behaviour Alignment
Nathan Gavenski, Matteo Leonetti and Odinaldo Rodrigues
State-of-the-art imitation learning from observation methods (ILfO) have recently made significant progress, but they still have some limitations: they need action-based supervised optimisation, assume that states have a single optimal action, and tend to apply teacher actions without full consideration of the actual environment state. While the truth may be out there in observed trajectories, existing methods struggle to extract it without supervision. In this work, we propose Unsupervised Imitation Learning from Observation (UfO) that addresses all of these limitations. UfO learns a policy through a two-stage process, in which the
agent first obtains an approximation of the teacher's true actions in the observed state transitions, and then refines the learned policy further by adjusting
agent trajectories to closely align them with the teacher's. Experiments we conducted in five widely used environments show that UfO not only outperforms the teacher and all other ILfO methods but also displays the smallest standard deviation. This reduction in standard deviation indicates better generalisation in unseen scenarios.
Read the full abstract ▾
Machine Learning · Cross-listed from math.NA
Sparse RBF Networks for PDEs and nonlocal equations: function space theory, operator calculus, and training algorithms
Zihan Shao, Konstantin Pieper and Xiaochuan Tian
This work presents a systematic analysis and extension of the sparse radial basis function network (SparseRBFnet) previously introduced for solving nonlinear partial differential equations (PDEs). Based on its adaptive-width shallow kernel network formulation, we further investigate its function-space characterization, operator evaluation, and computational
algorithm. We provide a unified description of the solution space for a broad class of radial basis functions (RBFs). Under mild assumptions, this space admits a characterization as a Besov space, independent of the specific kernel choice. We further demonstrate how the explicit kernel-based structure enables quasi-analytical evaluation of both differential and nonlocal operators, including fractional
Laplacians. On the computational end, we study the adaptive-width network and related three-phase training strategy through a comparison with variants concerning the modeling and algorithmic details. In particular, we assess the roles of second-order optimization, inner-weight training, network adaptivity, and anisotropic kernel parameterizations. Numerical experiments on high-order, fractional, and anisotropic PDE
benchmarks illustrate the empirical insensitivity to kernel choice, as well as the resulting trade-offs between accuracy, sparsity, and computational cost. Collectively, these results consolidate and generalize the theoretical and computational framework of SparseRBFnet, supporting accurate sparse representations with efficient operator evaluation and offering theory-grounded guidance for algorithmic and modeling choices.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.PF
GreenServ: Energy-Efficient Context-Aware Dynamic Routing for Multi-Model LLM Inference
Thomas Ziller, Shashikant Ilager, Alessandro Tundo et al. (+3 more)
Large
language models (LLMs) demonstrate remarkable capabilities, but their broad deployment is limited by significant computational resource demands, particularly energy consumption during inference. Static, one-model-fits-all inference strategies are often inefficient, as they do not exploit the diverse range of available models or adapt to varying query requirements. This paper presents GreenServ, a dynamic, context-aware routing framework that optimizes the trade-off between inference accuracy and energy efficiency. GreenServ extracts lightweight contextual features from each query, including task type, semantic cluster, and text complexity, and routes queries to the most suitable model from a heterogeneous pool, based on observed accuracy and energy usage. We employ a multi-armed bandit approach to learn adaptive routing policies online. This approach operates under partial feedback, eliminates the need for extensive offline calibration, and streamlines the integration of new models into the inference pipeline. We evaluated GreenServ across five
benchmark tasks and a pool of 16 contemporary open-access LLMs. Experimental results show that GreenServ consistently outperforms static (single-model) and random baselines. In particular, compared to random routing, GreenServ achieved a 22% increase in accuracy while reducing cumulative energy consumption by 31%. Finally, we evaluated GreenServ with RouterBench, achieving an average accuracy of 71.7% with a peak accuracy of 75.7%. All artifacts are open-source and available here: \href{https://github.com/TZData1/llm-inference-router}{GitHub}
Read the full abstract ▾
All authors Thomas Ziller, Shashikant Ilager, Alessandro Tundo, Ezio Bartocci, Leonardo Mariani, Ivona Brandic
Machine Learning · Cross-listed from cs.CV
OTI: A Model-free and Visually Interpretable Measure of Image Attackability
Jiaming Liang, Haowei Liu and Chi-Man Pun
Despite the tremendous success of neural networks, benign images can be corrupted by
adversarial perturbations to deceive these models. Intriguingly, images differ in their attackability. Specifically, given an attack configuration, some images are easily corrupted, whereas others are more resistant. Evaluating image attackability has important applications in active learning,
adversarial training, and attack enhancement. This
prompts a growing interest in developing attackability measures. However, existing methods are scarce and suffer from two major limitations: (1) They rely on a model proxy to provide prior knowledge (e.g., gradients or minimal perturbation) to extract model-dependent image features. Unfortunately, in practice, many task-specific models are not readily accessible. (2) Extracted features characterizing image attackability lack visual interpretability, obscuring their direct relationship with the images. To address these, we propose a novel Object Texture Intensity (OTI), a model-free and visually interpretable measure of image attackability, which measures image attackability as the texture intensity of the image's semantic object. Theoretically, we describe the principles of OTI from the perspectives of decision boundaries as well as the mid- and high-frequency characteristics of
adversarial perturbations. Comprehensive experiments demonstrate that OTI is effective and computationally efficient. In addition, our OTI provides the
adversarial machine learning community with a visual understanding of attackability.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CR
Reconstructing Training Data from Adapter-based Federated Large Language Models
Silong Chen, Yuchuan Luo, Guilin Deng et al. (+4 more)
Adapter-based
Federated Large
Language Models (FedLLMs) are widely adopted to reduce the computational, storage, and communication overhead of full-parameter fine-tuning for web-scale applications while preserving user
privacy. By freezing the backbone and training only compact low-rank adapters, these methods appear to limit gradient leakage and thwart existing Gradient Inversion Attacks (GIAs). Contrary to this assumption, we show that low-rank adapters create new, exploitable leakage channels. We propose the Unordered-word-bag-based Text Reconstruction (UTR) attack, a novel GIA tailored to the unique structure of adapter-based FedLLMs. UTR overcomes three core challenges: low-dimensional gradients, frozen backbones, and combinatorially large reconstruction spaces by: (i) inferring
token presence from
attention patterns in frozen layers, (ii) performing sentence-level inversion within the low-rank subspace of adapter gradients, and (iii) enforcing semantic coherence through constrained greedy decoding guided by
language priors. Extensive experiments across diverse models (GPT2-Large, BERT, Qwen2.5-7B) and
datasets (CoLA, SST-2, Rotten Tomatoes) demonstrate that UTR achieves near-perfect reconstruction accuracy (ROUGE-1/2 > 99), even with large batch size settings where prior GIAs fail completely. Our results reveal a fundamental tension between parameter efficiency and
privacy in FedLLMs, challenging the prevailing belief that lightweight adaptation inherently enhances security. Our code and data are available at https://github.com/shwksnshwowk-wq/GIA.
Read the full abstract ▾
All authors Silong Chen, Yuchuan Luo, Guilin Deng, Yi Liu, Min Xu, Shaojing Fu, Xiaohua Jia
Machine Learning · Cross-listed from cs.SD
EuleroDec: A Complex-Valued RVQ-VAE for Efficient and Robust Audio Coding
Luca Cerovaz, Michele Mancusi and Emanuele Rodolà
Audio codecs power discrete music generative modelling, music streaming and immersive media by shrinking PCM audio to bandwidth-friendly bit-rates. Recent works have gravitated towards processing in the spectral domain; however, spectrogram-domains typically struggle with phase modeling which is naturally complex-valued. Most frequency-domain neural codecs either disregard phase information or encode it as two separate real-valued channels, limiting spatial fidelity. This entails the need to introduce
adversarial discriminators at the expense of convergence speed and training stability to compensate for the inadequate representation power of the audio signal. In this work we introduce an end-to-end complex-valued RVQ-VAE audio codec that preserves magnitude-phase coupling across the entire analysis-quantization-synthesis pipeline and removes
adversarial discriminators and
diffusion post-filters. Without GANs or
diffusion we match or surpass much longer-trained baselines in-domain and reach SOTA out-of-domain performance. Compared to standard baselines that train for hundreds of thousands of steps, our model reducing training budget by an order of magnitude is markedly more compute-efficient while preserving high perceptual quality.
Read the full abstract ▾
Machine Learning
One-Shot Federated Clustering of Non-Independent Completely Distributed Data
Yiqun Zhang, Shenghong Cai, Zihua Yang et al. (+3 more)
Federated Learning (FL) that extracts data knowledge while protecting the
privacy of multiple clients has achieved remarkable results in distributed
privacy-preserving IoT systems, including smart traffic flow monitoring, smart grid load balancing, and so on. Since most data collected from edge devices are unlabeled, unsupervised
Federated Clustering (FC) is becoming increasingly popular for exploring pattern knowledge from complex distributed data. However, due to the lack of label guidance, the common Non-Independent and Identically Distributed (Non-IID) issue of clients have greatly challenged FC by posing the following problems: How to fuse pattern knowledge (i.e., cluster distribution) from Non-IID clients; How are the cluster distributions among clients related; and How does this relationship connect with the global knowledge fusion? In this paper, a more tricky but overlooked phenomenon in Non-IID is revealed, which bottlenecks the clustering performance of the existing FC approaches. That is, different clients could fragment a cluster, and accordingly, a more generalized Non-IID concept, i.e., Non-ICD (Non-Independent Completely Distributed), is derived. To tackle the above FC challenges, a new framework named GOLD (Global Oriented Local Distribution Learning) is proposed. GOLD first finely explores the potential incomplete local cluster distributions of clients, then uploads the distribution summarization to the server for global fusion, and finally performs local cluster enhancement under the guidance of the global distribution. Extensive experiments, including significance tests, ablation studies, scalability evaluations, qualitative results, etc., have been conducted to show the superiority of GOLD.
Read the full abstract ▾
All authors Yiqun Zhang, Shenghong Cai, Zihua Yang, Sen Feng, Yuzhu Ji, Haijun Zhang
Machine Learning · Cross-listed from stat.ML
"Rebuilding" Statistics in the Age of AI: A Town Hall Discussion on Culture, Infrastructure, and Training
David L. Donoho, Jian Kang, Xihong Lin et al. (+7 more)
This article presents the full, original record of the 2024 Joint Statistical Meetings (JSM) town
hall, "Statistics in the Age of AI," which convened leading statisticians to discuss how the field is evolving in response to advances in artificial intelligence, foundation models, large-scale empirical modeling, and data-intensive infrastructures. The town
hall was structured around open panel discussion and extensive audience Q&A, with the aim of eliciting candid, experience-driven perspectives rather than formal presentations or prepared statements. This document preserves the extended exchanges among panelists and audience members, with minimal editorial intervention, and organizes the conversation around five recurring questions concerning disciplinary culture and practices, data curation and "data work," engagement with modern empirical modeling, training for large-scale AI applications, and partnerships with key AI stakeholders. By providing an archival record of this discussion, the preprint aims to support transparency, community reflection, and ongoing dialogue about the evolving role of statistics in the data- and AI-centric future.
Read the full abstract ▾
All authors David L. Donoho, Jian Kang, Xihong Lin, Bhramar Mukherjee, Dan Nettleton, Rebecca Nugent, Abel Rodriguez, Eric P. Xing, Tian Zheng, Hongtu Zhu
Machine Learning
PEARL: Prototype-Enhanced Alignment for Label-Efficient Representation Learning with Deployment-Driven Insights from Digital Governance Communication Systems
Ruiyu Zhang, Lin Nie, Wai-Fung Lam et al. (+2 more)
In many deployed systems, new text inputs are handled by retrieving similar past cases, for example when routing and responding to citizen messages in digital governance platforms. When these systems fail, the problem is often not the
language model itself, but that the nearest neighbors in the
embedding space correspond to the wrong cases. Modern machine learning systems increasingly rely on fixed, high-dimensional
embeddings produced by large pretrained models and sentence encoders. In real-world deployments, labels are scarce, domains shift over time, and retraining the base encoder is expensive or infeasible. As a result, downstream performance depends heavily on
embedding geometry. Yet raw
embeddings are often poorly aligned with the local neighborhood structure required by nearest-neighbor
retrieval, similarity search, and lightweight classifiers that operate directly on
embeddings. We propose PEARL (Prototype-Enhanced Aligned Representation Learning), a label-efficient approach that uses limited supervision to softly align
embeddings toward class prototypes. The method reshapes local neighborhood geometry while preserving dimensionality and avoiding aggressive projection or collapse. Its aim is to bridge the gap between purely unsupervised post-processing, which offers limited and inconsistent gains, and fully supervised projections that require substantial labeled data. We evaluate PEARL under controlled label regimes ranging from extreme label scarcity to higher-label settings. In the label-scarce condition, PEARL substantially improves local neighborhood quality, yielding 25.7% gains over raw
embeddings and more than 21.1% gains relative to strong unsupervised post-processing, precisely in the regime where similarity-based systems are most brittle.
Read the full abstract ▾
All authors Ruiyu Zhang, Lin Nie, Wai-Fung Lam, Qihao Wang, Xin Zhao
Machine Learning
SpatialMath: Spatial Comprehension-Infused Symbolic Reasoning for Mathematical Problem-Solving
Ashutosh Bajpai, Akshat Bhandari, Akshay Nambi et al. (+1 more)
Multimodal Small-to-Medium sized
Language Models (MSLMs) have demonstrated strong capabilities in integrating visual and textual information but still face significant limitations in visual comprehension and mathematical
reasoning, particularly in
geometric problems with diverse levels of visual infusion. Current models struggle to accurately decompose intricate visual inputs and connect perception with structured
reasoning, leading to suboptimal performance. To address these challenges, we propose SpatialMath, a novel Spatial Comprehension-Infused Symbolic
Reasoning Framework designed to integrate spatial representations into structured symbolic
reasoning chains. SpatialMath employs a specialized perception module to extract spatially-grounded representations from visual diagrams, capturing critical
geometric structures and spatial relationships. These representations are then methodically infused into symbolic
reasoning chains, facilitating visual comprehension-aware structured
reasoning. To this end, we introduce MATHVERSE-PLUS, a novel
dataset containing structured visual interpretations and step-by-step
reasoning paths for
vision-intensive mathematical problems. SpatialMath significantly outperforms strong
multimodal baselines, achieving up to 10 percentage points improvement over supervised fine-tuning with data augmentation in
vision-intensive settings. Robustness analysis reveals that enhanced spatial representations directly improve
reasoning accuracy, reinforcing the need for structured perception-to-
reasoning pipelines in MSLMs.
Read the full abstract ▾
All authors Ashutosh Bajpai, Akshat Bhandari, Akshay Nambi, Tanmoy Chakraborty
Machine Learning
Automatic Stability and Recovery for Neural Network Training
Barak Or
Training modern neural networks is increasingly fragile, with rare but severe destabilizing updates often causing irreversible divergence or silent performance degradation. Existing optimization methods primarily rely on preventive mechanisms embedded within the optimizer, offering limited ability to detect and recover from instability once it occurs. We introduce a supervisory runtime stability framework that treats optimization as a controlled
stochastic process. By isolating an innovation signal derived from secondary measurements, such as validation probes, the framework enables automatic detection and recovery from destabilizing updates without modifying the underlying optimizer. We provide theoretical runtime safety guarantees that formalize bounded degradation and recovery. Our implementation incurs minimal overhead and is compatible with
memory-constrained training settings.
Read the full abstract ▾
Machine Learning
Unintended Memorization of Sensitive Information in Fine-Tuned Language Models
Marton Szep, Jorge Marin Ruiz, Georgios Kaissis et al. (+4 more)
Fine-tuning Large
Language Models (LLMs) on sensitive
datasets carries a substantial risk of unintended memorization and leakage of Personally Identifiable Information (PII), which can violate
privacy regulations and compromise individual safety. In this work, we systematically investigate a critical and underexplored vulnerability: the exposure of PII that appears only in model inputs, not in training targets. Using both synthetic and real-world
datasets, we design controlled extraction probes to quantify unintended PII memorization and study how factors such as
language, PII frequency, task type, and model size influence memorization behavior. We further
benchmark four
privacy-preserving approaches including differential
privacy, machine unlearning, regularization, and preference
alignment, evaluating their trade-offs between
privacy and task performance. Our results show that post-training methods generally provide more consistent
privacy-utility trade-offs, while differential
privacy achieves strong reduction in leakage in specific settings, although it can introduce training instability. These findings highlight the persistent challenge of memorization in fine-tuned LLMs and emphasize the need for robust, scalable
privacy-preserving techniques.
Read the full abstract ▾
All authors Marton Szep, Jorge Marin Ruiz, Georgios Kaissis, Paulina Seidl, Rüdiger von Eisenhart-Rothe, Florian Hinterwimmer, Daniel Rueckert
Machine Learning
LeanTutor: Towards a Verified AI Mathematical Proof Tutor
Manooshree Patel, Rayna Bhattacharyya, Thomas Lu et al. (+4 more)
This paper considers the development of an AI-based provably-correct mathematical proof tutor. While Large
Language Models (LLMs) allow seamless communication in natural
language, they are error prone.
Theorem provers such as Lean allow for provable-correctness, but these are hard for students to learn. We present a proof-of-concept system (LeanTutor) by combining the complementary strengths of LLMs and
theorem provers. LeanTutor is composed of three modules: (i) an autoformalizer/proof-checker, (ii) a next-step generator, and (iii) a natural
language feedback generator. To evaluate the system, we introduce PeanoBench, a
dataset of 371 Peano Arithmetic proofs in human-written natural
language and formal
language, derived from the Natural Numbers Game.
Read the full abstract ▾
All authors Manooshree Patel, Rayna Bhattacharyya, Thomas Lu, Arnav Mehta, Niels Voss, Narges Norouzi, Gireeja Ranade
Machine Learning
Identifying and Correcting Label Noise for Robust GNNs via Influence Contradiction
Wei Ju, Wei Zhang, Siyu Yi et al. (+6 more)
Graph Neural Networks (GNNs) have shown remarkable capabilities in learning from graph-structured data with various applications such as social analysis and bioinformatics. However, the presence of label noise in real scenarios poses a significant challenge in learning robust GNNs, and their effectiveness can be severely impacted when dealing with noisy labels on graphs, often stemming from annotation errors or inconsistencies. To address this, in this paper we propose a novel approach called ICGNN that harnesses the structure information of the graph to effectively alleviate the challenges posed by noisy labels. Specifically, we first design a novel noise indicator that measures the influence contradiction score (ICS) based on the graph
diffusion matrix to quantify the credibility of nodes with clean labels, such that nodes with higher ICS values are more likely to be detected as having noisy labels. Then we leverage the Gaussian mixture model to precisely detect whether the label of a node is noisy or not. Additionally, we develop a soft strategy to combine the predictions from neighboring nodes on the graph to correct the detected noisy labels. At last, pseudo-labeling for abundant unlabeled nodes is incorporated to provide auxiliary supervision signals and guide the model optimization. Experiments on
benchmark datasets show the superiority of our approach over competitive baselines in noisy label scenarios.
Read the full abstract ▾
All authors Wei Ju, Wei Zhang, Siyu Yi, Zhengyang Mao, Yifan Wang, Jingyang Yuan, Zhiping Xiao, Ziyue Qiao, Ming Zhang
Machine Learning · Cross-listed from cs.CV
ReflexSplit: Single Image Reflection Separation via Layer Fusion-Separation
Chia-Ming Lee, Yu-Fan Lin, Jin-Hui Jiang et al. (+3 more)
Single Image Reflection Separation (SIRS) disentangles mixed images into transmission and reflection layers. Existing methods suffer from transmission-reflection confusion under nonlinear mixing, particularly in deep decoder layers, due to implicit fusion mechanisms and inadequate multi-scale coordination. We propose ReflexSplit, a dual-stream framework with three key innovations. (1) Cross-scale Gated Fusion (CrGF) adaptively aggregates semantic priors, texture details, and decoder context across hierarchical depths, stabilizing gradient flow and maintaining feature consistency. (2) Layer Fusion-Separation Blocks (LFSB) alternate between fusion for shared structure extraction and differential separation for layer-specific disentanglement. Inspired by Differential
Transformer, we extend
attention cancellation to dual-stream separation via cross-stream subtraction. (3) Curriculum training progressively strengthens differential separation through depth-dependent initialization and epoch-wise warmup. Extensive experiments on synthetic and real-world
benchmarks demonstrate state-of-the-art performance with superior perceptual quality and robust generalization. Our code is available at https://github.com/wuw2135/ReflexSplit.
Read the full abstract ▾
All authors Chia-Ming Lee, Yu-Fan Lin, Jin-Hui Jiang, Yu-Jou Hsiao, Chih-Chung Hsu, Yu-Lun Liu
Machine Learning
Harnessing Reasoning Trajectories for Hallucination Detection via Answer-agreement Representation Shaping
Jianxiong Zhang, Bing Guo, Yuming Jiang et al. (+3 more)
Large
reasoning models (LRMs) often generate long, seemingly coherent
reasoning traces yet still produce incorrect answers, making
hallucination detection challenging. Although trajectories contain useful signals, directly using trace text or vanilla hidden states for detection is brittle: traces vary in form and detectors can overfit to superficial patterns rather than answer validity. We introduce Answer-agreement Representation Shaping (ARS), which learns detection-friendly trace-conditioned representations by explicitly encoding answer stability. ARS generates counterfactual answers through small latent interventions, specifically, perturbing the trace-boundary
embedding, and labels each perturbation by whether the resulting answer agrees with the original. It then learns representations that bring answer-agreeing states together and separate answer-disagreeing ones, exposing latent instability indicative of
hallucination risk. The shaped
embeddings are plug-and-play with existing
embedding-based detectors and require no human annotations during training. Experiments demonstrate that ARS consistently improves detection and achieves substantial gains over strong baselines. Code is available at: https://github.com/radiolab-ntu/ars_icml2026.
Read the full abstract ▾
All authors Jianxiong Zhang, Bing Guo, Yuming Jiang, Haobo Wang, Bo An, Sean Du
Machine Learning · Cross-listed from quant-ph
Bayesian quantum sensing using graybox machine learning
Akram Youssry, Stefan Todd, Patrick Murton et al. (+4 more)
Quantum sensors offer significant advantages over classical devices in spatial resolution and sensitivity, enabling transformative applications across materials science, healthcare, and beyond. Their practical performance, however, is often constrained by unmodelled effects, including noise, imperfect state preparation, and non-ideal control fields. In this work, we report the first experimental implementation of a graybox modelling strategy for a solid-state open
quantum system. The graybox framework integrates a physics-based system model with a data-driven description of experimental imperfections, achieving higher fidelity than purely analytical (whitebox) approaches while requiring fewer training resources than fully deep-learning blackbox models. We experimentally validate the method on the task of estimating a static
magnetic field using a single-
spin quantum sensor, performing
Bayesian inference with a graybox model trained on prior experimental data. Using roughly 10,000 training datapoints, the graybox model yields several orders of magnitude improvement in mean squared error over the corresponding physics-only model, as well as significant improvement over a blackbox model with comparable size. These results are broadly applicable to a wide range of
quantum sensing platforms, not limited to single-
spin systems, and are particularly valuable for real-time adaptive protocols, where model inaccuracies can otherwise lead to suboptimal control and degraded performance.
Read the full abstract ▾
All authors Akram Youssry, Stefan Todd, Patrick Murton, Muhammad Junaid Arshad, Nicholas Werren, Alberto Peruzzo, Cristian Bonato
Machine Learning · Cross-listed from cs.MA
Does Centralized Learning Help Embodied Agents Coordinate, and When Can a Tabular Study Tell?
Muhammad Ahmed Atif, Nehal Naeem Haji, Muhammad Ebad Atif et al. (+1 more)
Tabular studies are a common way to compare independent and centralized multi-
agent learners, because they avoid function approximation. A natural expectation is that centralized learning, which learns joint actions, helps
agents coordinate, especially when their bodies make solo success hard. Earlier work tests this only in parts: it measures coordination for one learning method, compares methods with the
agents' bodies fixed, or varies speed but reports only how well the team did. We test it in a predator-prey environment, where teammates cooperatively hunt together while two teams compete. We compare independent Q-learning (IQL) with central Q-learning (CQL) in all four predator and prey pairings, under limits on speed and stamina, on a 5x5 and an 8x8 grid trained for 200,000 episodes each, and check that the
agents learn on both. We measure coordination directly, with predictions for the main measurement fixed in advance. In this task, where one predator can capture prey alone, we find three things. First, CQL costs its
agents in both roles on both grids, and its predators coordinate no more than IQL's. Second, its one benefit comes from acting on the best joint action: this cut capture time by 42% on the small grid with faster prey and unlimited stamina, but helped nowhere on the larger grid, where the best joint action had mostly never been tried. Third, on the larger grid a tie-breaking default decided which method won, even though the
agents had passed a learning check. A tabular comparison can therefore be trusted only if it also checks how much of the state space the
agents covered and implements the methods alike. We propose simple checks for both.
Read the full abstract ▾
All authors Muhammad Ahmed Atif, Nehal Naeem Haji, Muhammad Ebad Atif, Mohammad Shahid Shaikh
Machine Learning
DREAM: Dual-Standard Semantic Homogeneity with Dynamic Optimization for Graph Learning with Label Noise
Yusheng Zhao, Jiaye Xie, Qixin Zhang et al. (+5 more)
Graph neural networks (GNNs) have been widely used in various graph machine learning scenarios. Existing literature primarily assumes well-annotated training graphs, while the reliability of labels is not guaranteed in real-world scenarios. Recently, efforts have been made to address the problem of graph learning with label noise. However, existing methods often (i) struggle to distinguish between reliable and unreliable nodes, and (ii) overlook the relational information embedded in the graph
topology. To tackle this problem, this paper proposes a novel method, Dual-Standard Semantic Homogeneity with Dynamic Optimization (DREAM), for reliable, relation-informed optimization on graphs with label noise. Specifically, we design a relation-informed dynamic optimization framework that iteratively reevaluates the reliability of each labeled node in the graph during the optimization process according to the relation of the target node and other nodes. To measure this relation comprehensively, we propose a dual-standard selection strategy that selects a set of anchor nodes based on both node proximity and graph
topology. Subsequently, we compute the semantic homogeneity between the target node and the anchor nodes, which serves as guidance for optimization. We also provide a rigorous theoretical analysis to justify the design of DREAM. Extensive experiments are performed on six graph
datasets across various domains under three types of graph label noise against competing baselines, and the results demonstrate the effectiveness of the proposed DREAM.
Read the full abstract ▾
All authors Yusheng Zhao, Jiaye Xie, Qixin Zhang, Weizhi Zhang, Xiao Luo, Zhiping Xiao, Philip S. Yu, Ming Zhang
Machine Learning · Cross-listed from cs.CL
Clustering-driven Memory Compression for On-device Large Language Models
Ondrej Bohdal, Pramit Saha, Umberto Michieli et al. (+2 more)
Large
language models (LLMs) often rely on user-specific
memories distilled from past interactions to enable personalized generation. A common practice is to concatenate these
memories with the input
prompt, but this approach quickly exhausts the limited context available in on-device LLMs. Compressing
memories by averaging can mitigate context growth, yet it frequently harms performance due to semantic conflicts across heterogeneous
memories. In this work, we introduce a clustering-based
memory compression strategy that balances context efficiency and personalization quality. Our method groups
memories by similarity and merges them within clusters prior to concatenation, thereby preserving coherence while reducing redundancy. Experiments demonstrate that our approach substantially lowers the number of
memory tokens while outperforming baseline strategies such as naive averaging or direct concatenation. Furthermore, for a fixed context budget, clustering-driven merging yields more compact
memory representations and consistently enhances generation quality.
Read the full abstract ▾
All authors Ondrej Bohdal, Pramit Saha, Umberto Michieli, Mete Ozay, Taha Ceritli
Machine Learning · Cross-listed from eess.SY
A new approach for combined model class selection and parameters learning for auto-regressive neural models
Corrado Sgadari, Alessio La Bella and Marcello Farina
This work introduces a novel approach for the joint selection of model structure and parameter learning for nonlinear dynamical systems identification. Focusing on a specific Recurrent Neural Networks (RNNs) family, i.e., Nonlinear Auto-Regressive with eXogenous inputs Echo State Networks (NARXESNs), the method allows to simultaneously select the optimal model class and learn model parameters from data through a new set-membership (SM) based procedure. The results show the effectiveness of the approach in identifying parsimonious yet accurate models suitable for control applications. Moreover, the proposed framework enables a robust training strategy that explicitly accounts for bounded measurement noise and enhances model robustness by allowing data-consistent evaluation of simulation performance during parameter learning, a process generally NP-hard for models with autoregressive components.
Read the full abstract ▾
Machine Learning
Data-driven Clustering and Merging of Adapters for On-device Large Language Models
Ondrej Bohdal, Taha Ceritli, Mete Ozay et al. (+4 more)
On-device large
language models commonly employ task-specific adapters (e.g., LoRAs) to deliver strong performance on downstream tasks. While storing all available adapters is impractical due to
memory constraints, mobile devices typically have sufficient capacity to store a limited number of these parameters. This raises a critical challenge: how to select representative adapters that generalize well across multiple tasks - a problem that remains unexplored in existing literature. We propose a novel method D2C for adapter clustering that leverages minimal task-specific examples (e.g., 10 per task) and employs an iterative optimization process to refine cluster assignments. The adapters within each cluster are merged, creating multi-task adapters deployable on resource-constrained devices. Experimental results demonstrate that our method effectively boosts performance for considered storage budgets.
Read the full abstract ▾
All authors Ondrej Bohdal, Taha Ceritli, Mete Ozay, Jijoong Moon, Kyeng-Hun Lee, Hyeonmok Ko, Umberto Michieli
Machine Learning · Cross-listed from cs.IR
UniGRec: Unified Generative Recommendation with Soft Identifiers for End-to-End Optimization
Jialei Li, Yang Zhang, Yimeng Bai et al. (+7 more)
Generative recommendation has recently emerged as a transformative paradigm that directly generates target items, surpassing traditional cascaded approaches. It typically involves two components: a tokenizer that learns item identifiers and a recommender trained on them. Existing methods often decouple tokenization from recommendation or rely on asynchronous alternating optimization, limiting full end-to-end
alignment. To address this, we unify the tokenizer and recommender under the ultimate recommendation objective via differentiable soft item identifiers, enabling joint end-to-end training. However, this introduces three challenges: training-inference discrepancy due to soft-to-hard mismatch, item identifier collapse from codeword usage imbalance, and collaborative signal deficiency due to an overemphasis on fine-grained
token-level semantics. To tackle these challenges, we propose UniGRec, a unified generative recommendation framework that addresses them from three perspectives. UniGRec employs Annealed Inference
Alignment during tokenization to smoothly bridge soft training and hard inference, a Codeword Uniformity Regularization to prevent identifier collapse and encourage codebook diversity, and a Dual Collaborative
Distillation mechanism that distills collaborative priors from a lightweight teacher model to jointly guide both the tokenizer and the recommender. Extensive experiments on real-world
datasets demonstrate that UniGRec consistently outperforms state-of-the-art baseline methods. Our codes are available at https://github.com/Jialei-03/UniGRec.
Read the full abstract ▾
All authors Jialei Li, Yang Zhang, Yimeng Bai, Shuai Zhu, Ziqi Xue, Xiaoyan Zhao, Dingxian Wang, Frank Yang, Andrew Rabinovich, Xiangnan He
Machine Learning
Active Hypothesis Testing for Correlated Combinatorial Anomaly Detection
Zichuan Yang and Yiming Xing
We study the problem of identifying an anomalous subset of streams under correlated noise, motivated by monitoring and security in cyber-physical systems. This problem can be viewed as a form of combinatorial pure exploration, where each stream plays the role of an arm and measurements must be allocated sequentially under uncertainty. Existing combinatorial bandit and hypothesis testing methods typically assume independent observations and fail to exploit correlation for efficient measurement design. We propose ECC-AHT, an adaptive
algorithm that selects continuous, constrained measurements to maximize Chernoff information between competing hypotheses, enabling active noise cancellation through differential sensing. ECC-AHT achieves optimal sample complexity guarantees and significantly outperforms state-of-the-art baselines in both synthetic and real-world correlated environments. The code is available on https://github.com/VincentdeCristo/ECC-AHT
Read the full abstract ▾
Machine Learning
Efficient Dilated Squeeze and Excitation Neural Operator for Differential Equations
Prajwal Chauhan, Salah Eddine Choutri and Saif Eddin Jabari
Fast and accurate surrogates for physics-driven partial differential equations (PDEs) are essential in fields such as aerodynamics, porous media design, and flow control. However, many
transformer-based models and existing neural operators remain parameter-heavy, resulting in costly training and sluggish deployment. We propose D-SENO (Dilated Squeeze-Excitation Neural Operator), a lightweight operator learning framework for efficiently solving a wide range of PDEs, including airfoil potential flow, Darcy flow in porous media, pipe Poiseuille flow, and incompressible Navier Stokes vortical fields. D-SENO combines dilated convolution (DC) blocks with squeeze-and-excitation (SE) modules to jointly capture wide receptive fields and dynamics alongside channel-wise
attention, enabling both accurate and efficient PDE inference. Carefully chosen dilation rates allow the receptive field to focus on critical regions, effectively modeling long-range physical dependencies. Meanwhile, the SE modules adaptively recalibrate feature channels to emphasize dynamically relevant scales. Our model achieves training speed of up to approximately $20\times$ faster than standard
transformer-based models and neural operators, while also surpassing (or matching) them in accuracy across multiple PDE
benchmarks. Ablation studies show that removing the SE modules leads to a slight drop in performance.
Read the full abstract ▾
Machine Learning
GO-OSC and VASH: Geometry-Aware Representation Learning for Early Degradation Detection in Oscillatory Systems
Vashista Nobaub
Early-stage degradation in oscillatory systems often manifests as
geometric distortions of the dynamics, such as phase jitter, frequency drift, or loss of coherence, long before changes in signal energy are detectable. In this regime, classical energy-based diagnostics and unconstrained learned representations are structurally insensitive, leading to delayed or unstable detection. We introduce GO-OSC, a geometry-aware representation learning framework for oscillatory time series that enforces a canonical and identifiable latent parameterization, enabling stable comparison and aggregation across short, unlabeled windows. Building on this representation, we define a family of invariant linear
geometric probes that target degradation-relevant directions in latent space. We provide theoretical results showing that under early phase-only degradation, energy-based statistics have zero first-order detection power, whereas
geometric probes achieve strictly positive sensitivity. Our analysis characterizes when and why linear probing fails under non-identifiable representations and shows how canonicalization restores statistical detectability. Experiments on synthetic
benchmarks and real vibration
datasets validate the theory, demonstrating earlier detection, improved data efficiency, and robustness to operating condition changes.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CR
Res-MIA: A Training-Free Resolution-Based Membership Inference Attack on Federated Learning Models
Mohammad Zare and Pirooz Shamsinejadbabaki
Membership inference attacks (MIAs) pose a serious threat to the
privacy of machine learning models by allowing adversaries to determine whether a specific data sample was included in the training set. Although
federated learning (FL) is widely regarded as a
privacy-aware training paradigm due to its decentralized nature, recent evidence shows that the final global model can still leak sensitive membership information through black-box access. In this paper, we introduce Res-MIA, a novel training-free and black-box membership inference attack that exploits the sensitivity of deep models to high-frequency input details. Res-MIA progressively degrades the input resolution using controlled downsampling and restoration operations, and analyzes the resulting confidence decay in the model's predictions. Our key insight is that training samples exhibit a significantly steeper confidence decline under resolution erosion compared to non-member samples, revealing a robust membership signal. Res-MIA requires no shadow models, no auxiliary data, and only a limited number of forward queries to the target model. We evaluate the proposed attack on a
federated ResNet-18 trained on CIFAR-10, where it consistently outperforms existing training-free baselines and achieves an AUC of up to 0.88 with minimal computational overhead. These findings highlight frequency-sensitive overfitting as an important and previously underexplored source of
privacy leakage in
federated learning, and emphasize the need for
privacy-aware model designs that reduce reliance on fine-grained, non-robust input features.
Read the full abstract ▾