paperlist.ai

AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

arXiv

INTRYGUE: Induction-Aware Entropy Gating for Reliable RAG Uncertainty Estimation

Alexandra Bazarova, Andrei Volodichev, Daria Kotova, Alexey Zaytsev

While retrieval-augmented generation (RAG) significantly improves the factual reliability of LLMs, it does not eliminate hallucinations, so robust uncertainty quantification (UQ) remains essential. In this paper, we reveal that standard entropy-based UQ methods often fail in RAG settings due to a mechanistic paradox. An internal "tug-of-war" inherent to context utilization appears: while induction heads promote grounded responses by copying the correct answer, they collaterally trigger the previously established "entropy neurons". This interaction inflates predictive entropy, causing the model to signal false uncertainty on accurate outputs. To address this, we propose INTRYGUE (Induction-Aware Entropy Gating for Uncertainty Estimation), a mechanistically grounded method that gates predictive entropy based on the activation patterns of induction heads. Evaluated across four RAG benchmarks and six open-source LLMs (4B to 13B parameters), INTRYGUE consistently matches or outperforms a wide range of UQ baselines. Our findings demonstrate that hallucination detection in RAG benefits from combining predictive uncertainty with interpretable, internal signals of context utilization.

arXiv

AutoMOOSE: An Agentic AI for Autonomous Phase-Field Simulation

Sukriti Manna, Henry Chan, Subramanian K. R. S. Sankaranarayanan

Phase-field modeling links thermodynamics and kinetics to microstructural evolution, but multiphysics frameworks such as MOOSE require expertise to construct inputs, manage campaigns, diagnose failures, and validate results. We introduce AutoMOOSE, an open-source multi-agent framework that orchestrates the simulation lifecycle from a single natural-language prompt. Six specialized agents--Architect, Input Writer, Runner, Reviewer, Visualization, and a physics-grounded Skeptic--generate, execute, analyze, and adversarially test simulations against conservation laws, asymptotic limits, and scaling relations. We validate AutoMOOSE in two domains: non-conserved copper grain growth governed by Allen-Cahn dynamics and conserved Fe-Cr spinodal decomposition governed by Cahn-Hilliard dynamics. On a prospectively specified 25-task grain-growth benchmark spanning temperature, grain count, resolution, and model formulation, AutoMOOSE generates valid inputs for all 25 tasks, completes 19 simulations with measurable coarsening, and delivers 15 results that satisfy Burke-Turnbull kinetics ($R^2 \geq 0.90$) and survive Skeptic falsification. The failures trace to two identifiable generator defects. An ensemble of 1000 simulations recovers the prescribed activation energy to within 1% after finite-size extrapolation ($Q_\infty = 0.228$ eV versus $0.230$ eV). In the conserved domain, an agent-generated CALPHAD-based Fe-Cr simulation limits relative mass drift to $6.3 \times 10^{-6}$, dissipates free energy to a stable plateau, and evolves toward the equilibrium tie-line. Controlled ablations show that the full pipeline converts stochastic raw generation--3-7 successful tasks out of 8 across repeated trials--into consistent success on all 8 tasks. AutoMOOSE therefore provides a practical route from natural-language specification to reproducible, physics-validated phase-field simulation campaigns.

arXiv

Deterministic Preprocessing and Interpretable Fuzzy Banding for Cost-per-Student Reporting from Extracted Records

Shane Lee, Stella Ng

Administrative extracts are often exchanged as spreadsheets and may be read as reports in their own right during budgeting, workload review, and governance discussions. When an exported workbook becomes the reference snapshot for such decisions, the transformation can be checked by recomputation against a clearly identified input. A deterministic, rule-governed, file-based workflow is implemented in cad_processor.py. The script ingests a Casual Academic Database (CAD) export workbook and aggregates inclusive on-costs and student counts into subject-year and school-year totals, from which it derives cost-per-student ratios. It writes a processed workbook with four sheets: Processing Summary (run record and counters), Trend Analysis (schoolyear cost-per-student matrix), Report (wide subject-level table), and Fuzzy Bands (per-year anchors, membership weights, and band labels). The run record includes a SHA-256 hash of the input workbook bytes to support snapshot-matched recomputation. For within-year interpretation, the workflow adds a simple fuzzy banding layer that labels finite, positive school-year cost-per-student values as Low, Medium, or High. The per-year anchors are the minimum, median, and maximum of the finite, positive ratios. Membership weights are computed using left-shoulder, triangular, and right-shoulder functions, with deterministic tie-breaking in a fixed priority order (Medium, then Low, then High). These weights are treated as decision-support signals rather than probabilities. A worked example provides a reproducible calculation of a band assignment from the reported anchors and ratios. Supplementary material includes a claim-to-evidence matrix, a reproducibility note, and a short glossary that links selected statements to code and workbook artefacts.

arXiv

SAGEO Arena: A Realistic Environment for Evaluating Search-Augmented Generative Engine Optimization

Sunghwan Kim, Wooseok Jeong, Serin Kim, Sangam Lee, Dongha Lee

Search-Augmented Generative Engines (SAGE) have emerged as a new paradigm for information access, bridging web-scale retrieval with generative capabilities to deliver synthesized answers. This shift has fundamentally reshaped how web content gains exposure online, giving rise to Search-Augmented Generative Engine Optimization (SAGEO), the practice of optimizing web documents to improve their visibility in AI-generated responses. Despite growing interest, no evaluation environment currently supports comprehensive investigation of SAGEO. Specifically, existing benchmarks lack end-to-end visibility evaluation of optimization strategies, operating on pre-determined candidate documents that abstract away retrieval and reranking preceding generation. Moreover, existing benchmarks discard structural information (e.g., schema markup) present in real web documents, overlooking the rich signals that search systems actively leverage in practice. Motivated by these gaps, we introduce SAGEO Arena, a realistic and reproducible environment for stage-level SAGEO analysis. Our objective is to jointly target search-oriented optimization (SEO) and generation-centric optimization (GEO). To achieve this, we integrate a full generative search pipeline over a large-scale corpus of web documents with rich structural information. Our findings reveal that existing approaches remain largely impractical under realistic conditions and often degrade performance in retrieval and reranking. We also find that structural information helps mitigate these limitations, and that effective SAGEO requires tailoring optimization to each pipeline stage. Overall, our benchmark paves the way for realistic SAGEO evaluation and optimization beyond simplified settings.

arXiv

Optimizing Spectral Prediction in MXene-Based Metasurfaces Through Multi-Channel Spectral Refinement and Savitzky-Golay Smoothing

Shujaat Khan, Waleed Iqbal Waseer, Muhammad Shahid Jabbar

The prediction of electromagnetic spectra for MXene-based solar absorbers, where MXenes are a family of two-dimensional transition metal carbides and nitrides, is a computationally intensive task traditionally addressed using full-wave solvers. This study introduces an efficient deep learning framework incorporating transfer learning, multi-channel spectral refinement, and Savitzky-Golay smoothing to accelerate and enhance spectral prediction accuracy. The proposed architecture leverages a pretrained MobileNet version 2 model, fine-tuned to predict 102-point absorption spectra from ($64\times64$) metasurface designs. Additionally, the multi-channel spectral refinement module processes the feature map through multiple convolutional channels, enhancing feature extraction, while Savitzky-Golay smoothing mitigates high-frequency noise. Experimental evaluations demonstrate that the proposed model significantly outperforms baseline convolutional neural network and deformable convolutional neural network models, achieving an average root mean squared error of 0.0227, coefficient of determination ($R^2$) of 0.9563, and peak signal-to-noise ratio of 33.10 decibels. The proposed framework presents a scalable and computationally efficient alternative to conventional solvers, positioning it as a viable candidate for rapid spectral prediction in nanophotonic design workflows.

arXiv

Evaluating Useful Surrogate Models for Configuration Tuning Beyond Accuracy: A Fitness Landscape Analysis Perspective

Pengzhou Chen, Hongyuan Liang, Tao Chen

To efficiently tune configuration for better software system performance (e.g., latency) at the deployment and maintenance stage, many tuners have leveraged a surrogate model to expedite the process instead of solely relying on the profoundly expensive system measurement. As such, it is naturally believed that we need more accurate models. However, the fact of "accuracy can lie"-a somewhat surprising finding from prior work-has left us many unanswered questions regarding what role the surrogate model plays in configuration tuning. This paper provides the very first systematic exploration and discussion, together with a resolution proposal, to disclose the many faces of useful surrogate models for configuration tuning beyond accuracy, through the novel perspective of fitness landscape analysis. We present a theory as an alternative to accuracy for assessing the model usefulness in tuning, based on which we conduct an extensive empirical study involving up to 27,000 cases. Drawing on the above, we propose Model4Tune, an automated predictive tool that estimates which model-tuner pairs are the best for an unforeseen system without expensive tuner profiling. Our results suggest that Model4Tune, as one of the first of its kind, performs significantly better than random guessing in 79%-82% of the cases, hence greatly mitigating the required efforts in engineering configuration for software systems. Our results not only shed light on the possible future research directions but also offer a practical resolution that can assist practitioners in evaluating the most useful model for configuration tuning.

arXiv

Learning to Walk With Less: A Dyna-Style Approach to Quadrupedal Locomotion

Francisco Affonso, Felipe Tommaselli, João H. Aléssio, Vivian S. Medeiros, Mateus V. Gasparino, Girish Chowdhary, Marcelo Becker

Traditional on-policy reinforcement learning (RL) controllers for quadrupedal locomotion often suffer from low data efficiency, requiring millions of interactions with simulated environments to achieve stable control. We integrate model-based techniques that improve sample efficiency by augmenting PPO rollouts with synthetic data in a Dyna-style framework. Our method employs a learned transition model to generate short-horizon synthetic tails for each trajectory, anchored by physics-based simulation to preserve stability. A predefined scheduling strategy gradually integrates synthetic transitions, preventing model usage during early training stages when prediction accuracy is low. Through extensive ablation studies, we analyze how varying data parameters influence PPO's learning behavior. Finally, we validate our method in simulation on a Unitree Go1 robot, reaching convergence with substantially fewer simulation steps (19.64M vs. 27.53M) and a 12.24% reduction in wall-clock training time, without compromising policy performance or convergence. Cross-platform experiments on ANYmal and Unitree Go2 further confirm the framework's ability to learn high-dimensional locomotion control with substantially reduced simulation experience, despite reward trade-offs on complex morphologies.

arXiv

Serious Games: Human-AI Interaction, Evolution, and Coevolution

Nandini Doreswamy, Louise Horstmanshof

The serious games between humans and AI have only just begun. Evolutionary Game Theory (EGT) models the competitive and cooperative strategies of biological entities. EGT could help predict the potential evolutionary equilibrium of humans and AI. The objective of this work was to examine EGT models relevant to human-AI interaction, evolution, and co-evolution. Of thirteen EGT models considered, three were examined: the Hawk-Dove Game, Iterated Prisoner's Dilemma, and the War of Attrition. This selection was based on the widespread acceptance and clear relevance of these models to potential human-AI evolutionary dynamics and co-evolutionary trajectories. The Hawk-Dove Game predicts balanced mixed-strategy equilibria based on the costs of conflict. Iterated Prisoner's Dilemma suggests that repeated interaction may lead to cognitive co-evolution. The War of Attrition suggests that competition for resources may result in strategic co-evolution, asymmetric equilibria, and conventions on sharing resources. Each model was examined from the perspective of human and AI decision-making, from psychological and biological perspectives, and from an AI viewpoint. AI is being shaped by human input and is evolving in response to it. So too, neuroplasticity allows the human brain to evolve in response to stimuli. If humans and AI converge in future, what might be the result of human neuroplasticity combined with an ever-evolving AI? There are profound ethical and cognitive implications. EGT may provide a suitable framework to understand and predict human-AI interaction, evolution, and co-evolution. However, future research should extend beyond EGT and explore additional frameworks, empirical validation methods, and interdisciplinary perspectives. In the spirit of further exploration, an illustrative computational simulation is provided.

arXiv

A primer on optimal transport for causal inference with observational data

Florian F Gunsilius

The theory of optimal transportation has developed into a powerful and elegant framework for comparing probability distributions, with wide-ranging applications in all areas of science. The fundamental idea of analyzing probabilities by comparing their underlying state space naturally aligns with the core idea of causal inference, where understanding and quantifying counterfactual states is paramount. Despite this intuitive connection, explicit research at the intersection of optimal transport and causal inference is only beginning to develop. Yet, many foundational models in causal inference have implicitly relied on optimal transport principles for decades, without recognizing the underlying connection. Therefore, the goal of this review is to offer an introduction to the surprisingly deep existing connections between optimal transport and the identification of causal effects with observational data -- where optimal transport is not just a set of potential tools, but actually builds the foundation of model assumptions. As a result, this review is intended to unify the language and notation between different areas of statistics, mathematics, and econometrics, by pointing out these existing connections, and to explore novel problems and directions for future work in both areas derived from this realization.

arXiv

RIS-Aided mmWave Localization Under Cross-Link Interference via Beam-Domain ML Fingerprinting

Md Tarek Hassan, Dmitry Zelenchuk, Muhammad Ali Babar Abbasi

Accurate user equipment (UE) localization is critical for beam management in reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mmWave) based sixth-generation (6G) networks, especially if the direct base-station-UE links are unavailable. This paper proposes a beam-domain fingerprint framework that maps the received signal-to-noise ratio (SNR) across a small set of predefined RIS reflection states to the UE azimuth angle and range, without requiring channel state information (CSI). Crucially, we extend the framework to a realistic interference-impaired scenario in which a nearby cross-link interferer (CLI) corrupts the clean SNR fingerprint, yielding a signal-to-interference-plus-noise ratio (SINR) fingerprint; an interference-to-noise ratio (INR)-constrained calibration strategy keeps the interference level physically interpretable. Four machine-learning (ML) regressors are evaluated under both conditions. Simulation results at 28 GHz with a 20x20 RIS show that k-nearest neighbors (KNN) achieves the lowest angle MAE of 0.37 degrees and range MAE of 4 cm under clean conditions, rising to 1.4 degrees and 7.6 cm under interference. A key finding is that interference degrades angle estimation substantially more than range estimation across all models, a consequence of the asymmetric encoding of location information in the beam-domain fingerprint.

arXiv

Towards a Theoretical Understanding of Two Tower Recommendation Models

Amit Kumar Jaiswal

Production-grade recommender systems rely heavily on a large-scale corpus used by online media services, including Netflix, Pinterest, and Amazon. These systems enrich recommendations by learning users' and items' embeddings projected in a low-dimensional space with two tower models (two deep neural networks), which facilitate their embedding constructs to predict users' feedback associated with items. Despite its popularity for recommendations, its theoretical behaviors remain comprehensively unexplored. We study the asymptotic behaviors of the two tower model applied in two-stage recommenders that entail a strong convergence to the optimal recommender system. We establish certain theoretical properties and statistical assurance of the two tower recommender. In addition to asymptotic behaviors, we demonstrate that recommendation with two tower architecture attains faster convergence by relying on the intrinsic dimensions of the input features. Finally, we show numerically that the two tower recommender enables encapsulating the impacts of items' and users' attributes on ratings, resulting in better performance compared to existing methods conducted using synthetic and real-world data experiments.

arXiv

Post-Grokking Collapse at the Representation-Readout Interface in Muon-Trained Transformers

Ali Janati, Kaoutar El Maghraoui, Andrei Kanavalau, Anass Belfatmi

Under the standard split, Muon gets hidden matrices and AdamW embeddings/output head. Muon groks modular addition faster, but its solutions do not hold. All nine configurations on $(a+b) \bmod 113$ grok and later lose generalization. Across five seeds the selected AdamW reference falls below threshold on four, reaching 27.59%. Instability persists across two moduli, two widths, two training fractions, subtraction, and depth. The failure arises at the representation-readout interface, identified only jointly up to an invertible map unselected by the loss. After solving the training set, the gradient falls to order $10^{-6}$ and the optimizers respond differently: step-size elasticity is -0.03 for Muon versus +1.5 for AdamW, and the Muon group moves 8.0 times faster per parameter. From bit-identical states, freezing either group prevents failure. Freezing embeddings/readout removes it in five runs over 451,400 post-grokking steps and five paired seeds: unfrozen arms record 137-321 sub-threshold evaluations, frozen arms none. Removing Muon's normalization and orthogonalization is no substitute: it collapses representation from 326 effective conjugate pairs to 4, shows no recurrent collapse, and fails terminally. Fourier filtering separates circuit failure from masking. Across 43 checkpoints over five seeds and three regimes, the task-aligned family reaches exactly 100% alone. In circuit failure it no longer solves the task; in masking it remains perfect while the full model reaches 45.85%, giving a positive margin on every example, including errors, but being outvoted by a near-equal adversarial remainder. Rescaling it restores 99.9%; grokking is the same condition resolving upward. The task selects the family, swapping $(k,k)$ for $(k,-k)$ under subtraction. Across an abrupt collapse, standard Fourier support is unchanged and the power-distribution cosine remains 0.9899.

arXiv

Wasserstein Policy Gradient for Entropy-Regularized Linear-Quadratic Control

Zhaoyu Zhu, Rui Gao, Shuang Li

Wasserstein policy gradient (WPG) updates state-conditional action laws by transport in the action space. We study entropy-regularized discounted linear-quadratic (LQ) control. A Bellman verification argument shows that the unrestricted problem has a linear-Gaussian optimal policy, and the discounted-occupancy-weighted statewise Wasserstein gradient is tangent to this policy class. WPG therefore reduces exactly to a finite-dimensional ODE for the feedback gain and action covariance. We prove that this ODE is globally well posed and converges exponentially from every admissible initialization. For each fixed LQ problem, the exponent has a positive limit as the entropy temperature tends to zero and contains no perturbative factor of the form $\exp(-c/τ)$, while retaining the usual dependence on the conditioning of the control problem.

arXiv

Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits

Elena Dumitrescu, Gert Lek, Lydia Y. Chen, Jérémie Decouchant

Diffusion Large Language Models (DLLMs) replace autoregressive next-token prediction with iterative parallel denoising, yet their internal safety mechanisms remain poorly understood. In this work, we investigate DLLMs both as targets and as adversaries, exposing mechanistic vulnerabilities in diffusion-based alignment. We first show that safety alignment in DLLMs remains sparse and transferable across architectures. DLLMs initialized from autoregressive predecessors inherit the same mechanistic safety footprint as their source models, enabling transfer attacks via direct safety neuron mapping and pruning. Self-pruning increases attack success rates (ASR) from 2.6% to 73.8% on LLaDA and from 1.9% to 86.6% on Dream, while transfer pruning from Qwen2.5 increases ASR from 1.9% to 73.2% on Dream and from 7.0% to 86.3% on Fast-dLLM. Building on these findings, we introduce SN-Guided Diffusion, a fully offline black-box jailbreak framework that steers the diffusion process away from safety-triggering regions using a weighted safety neuron loss, which achieves near-perfect prompt separability (AUROC = 1.0 for benign-vs-jailbreak discrimination). Across multiple open and proprietary targets, our method achieves a transfer ASR of up to 77.1% on Llama-3-8B-Instruct, 86.9% on Qwen2.5-7B-Instruct, and 74.3% against Gemini-2.5-Flash-Lite, while requiring only 20 generation episodes per prompt. Compared to prior jailbreaking frameworks, our method achieves competitive transferability with orders-of-magnitude lower generation cost. Our codebase is available at https://github.com/ellyoana/sn-guided-diffusion.

arXiv

Beyond Myopic World Models: Long-Horizon End-to-End Training for Direct Future Prediction

Xinyi Li, zaishuo xia, Chenjie Hao, Yubei Chen

World models are expected to support imagination over extended temporal horizons, yet most are still trained through local few-step prediction objectives and deployed by recursively rolling out their own predictions. This creates a fundamental mismatch: few-step losses optimize local transition fidelity, while long-horizon prediction depends on how errors and gradients propagate through the entire trajectory. As a result, transitions with different downstream influence on the endpoint are treated uniformly during training, and small local errors are amplified through recursive inference. We argue that long-horizon accuracy is better achieved by optimizing directly, through an end-to-end endpoint prediction objective. To instantiate this paradigm, we introduce the Direct Prediction World Model (DPWM), a non-recursive architecture that compresses an action sequence of arbitrary length into a single embedding and predicts the endpoint observation in a single forward pass. This design avoids recurrent rollout in both prediction and gradient propagation, making long-horizon end-to-end training practical at horizons where unrolled autoregressive training becomes unstable. Empirically, DPWM substantially improves long-horizon endpoint prediction over recursive world-model baselines on continuous-control and pixel-based benchmarks, with larger gains as the prediction horizon increases. We further show that recurrent baselines benefit similarly when retrained with the same long-horizon endpoint objective, supporting our central claim that the training objective, rather than the particular backbone choice, is the main driver of long-horizon prediction accuracy. Our results suggest that world models can benefit from being trained and evaluated at the temporal scales where they are ultimately used, shifting the focus from local transition modeling toward long-horizon predictive accuracy.

arXiv

Beyond Post-Hoc Temperature Scaling: Bilevel Optimization for LLM Calibration

Ruochen Jin, Zhanliang Wang, Zongyu Dai, Jiancong Xiao, Bojian Hou

Preference alignment often makes large language models (LLMs) overconfident and poorly calibrated. Traditional post-hoc temperature scaling is inherently domain-dependent: a temperature fitted on one domain does not generalize across domains. This motivates us to modify model parameters during training to improve calibration. We propose maximizing the entropy of predictive distributions as the calibration objective, which directly targets overconfidence by discouraging overly concentrated predictions. Inspired by temperature scaling, we realize this through a bilevel optimization formulation, where the lower level trains the model under a parametric loss and the upper level selects loss hyperparameters to maximize entropy. To make the framework practical at LLM scale, we adopt an efficient first-order approximation that avoids explicit second-order computation. Across both multiple-choice and open-ended generative question answering, experiments demonstrate that our method yields well-calibrated LLMs with particular advantages in out-of-domain generalization.

arXiv

DynaCrys: Crystal Generation with Dynamic Space-Group Diffusion

Zhuotao Jin, Xiaoyun Wang, Nicholas Brawand, Roman Zubatyuk, Atul Thakur, Eric Qu, Boris Kozinsky, Justin Smith

The search for new crystalline materials spans an enormous compositional and structural space. Generating candidates in this space requires jointly modeling discrete crystallographic symmetry, elemental composition, and continuous geometry. We introduce DynaCrys, a generative model for crystals in which the space group co-evolves with Wyckoff occupations and elements through a coupled symbolic diffusion process. The structured space-group transitions follow crystallographic group-subgroup relations. As the space group changes, a shared, pretrained symmetry codebook provides both the legality-constrained stochastic decoder and the symmetry-constrained crystal-geometry model with a common representation of the corresponding Wyckoff vocabulary. Across large-scale evaluations using two independent relaxation-and-evaluation engines, DynaCrys achieves best-in-class performance in symmetry-aware discovery of stable, unique, and novel crystals, both overall and under the additional requirement of nontrivial post-relaxation symmetry. It also enables fast sampling while generating structures with consistently low relaxation-induced structural displacements.

arXiv

Uncovering expert objectives in production planning via inverse optimization: An industrial case study

Shivi Dixit, Rishabh Gupta, Adam Kelloway, John Wassick, Qi Zhang

Production planning in the manufacturing industry often relies on the use of optimization models, but defining an appropriate objective function can be a challenge. In practice, planners must balance competing goals, manage uncertainty, and account for qualitative business preferences that are difficult to quantify. As a result, many optimization models fail to match expert behavior, limiting trust and adoption. In this work, we propose a data-driven inverse optimization framework to infer the objective function implicitly captured in expert planners' decisions. We formulate the production planning problem as a mixed-integer linear program, where the unknown objective function is represented as a weighted sum of hypothesized cost terms. A suboptimality-loss-based inverse optimization method is then applied to learn the objective weights from historical production plans. The proposed approach is applied to a real industrial case provided by Dow, where the inferred weights reveal that avoiding inventory shortages and maintaining consistent cycle lengths dominate the planners' decision-making. Time- and product-dependent extensions further improve predictive accuracy and uncover evolving priorities. Expert interviews confirm the practical validity of these insights. Overall, this study shows that inverse optimization can transform tacit human expertise into interpretable models, enabling more accurate and trusted decision-support tools for complex industrial systems.

arXiv

FedDOSE: Federated Learning Framework Decomposing Site Effects for Modeling Brain Dynamic Functional Connectivity

Deepank Girish, Yi Hao Chan, Yubin Zheng, Sukrit Gupta, Jagath C. Rajapakse

Functional Magnetic Resonance Imaging ( fMRI ) data are often pooled into collaborative multi-site consortia, as deep learning models for analyses require large datasets to generalize well. While Federated Learning (FL) offers a privacy-preserving paradigm for collaborative training, standard approaches continue to struggle with statistical heterogeneity. In particular, site differences pose a key challenge in multi-site data settings. Additionally, existing FL approaches for fMRI rely on static Functional Connectivity ( FC), omitting dynamic information in brain networks. To address this, we propose FedDOSE, a novel framework that explicitly decomposes site differences for analysis of dynamic FC (dFC). FedDOSE introduces a Modularity-Guided Tucker Decomposition block to encode high-dimensional dFC tensors and capture modular-level spatio-temporal patterns efficiently. Class-specific prototypes are generated across all sites and subsequently aligned at the global level by using a combination of Optimal Transport (OT) barycenter formulation and Procrustes analysis. Extensive experiments for diagnosing Autism Spectrum Disorder (ASD) and Attention-Deficit Hyperactivity Disorder (ADHD) on three multi-site resting-state fMRI datasets: ABIDE-I, ABIDE-II, and ADHD-200, demonstrate that FedDOSE outperforms state-of-the-art methods in ASD and ADHD detection. Our results highlight its effectiveness in learning robust representations from multi-site datasets for reliable analysis.

arXiv

Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations

Ioannis Ziogas, Ensieh Khazaei, Bilal Taha, Aamna Al Shehhi, Ahsan H. Khandoker, Leontios J. Hadjileontiadis, Dimitrios Hatzinakos

Learning disentangled representations is a key requirement for developing versatile, general-purpose, and sustainable models in multi-modal wearable computing. However, existing approaches do not operate as full-stack wearable processors, i.e., they do not simultaneously address task-specific classification performance, disentangled and interpretable representation learning, fusion, and generative modeling of highly heterogeneous multi-modal time series. To address this gap, we introduce Omni-modal Variational Decomposition Autoencoders (OmniDecVAEs), a framework that efficiently learns multi-purpose representations in a unified and scalable manner from arbitrarily many modalities. OmniDecVAEs extend DecVAEs by learning modality-conditioned time-frequency latent subspaces through a multi-view self-supervised decomposition loss and a shared asymmetric autoencoder (AE) architecture. Results on a challenging omni-modal human activity recognition (HAR) setting with up to thirty modalities, demonstrate the ability of OmniDecVAEs to learn full-stack wearable representations. When compared to transformer-based and VAE-based methods, OmniDecVAEs full-stack disentangled representation properties lead to accuracy improvements of 1.01% and 6.75% in activity and identity recognition, respectively. Furthermore, OmniDecVAEs synthesize realistic omni-modal time-frequency data that manifest with enhanced reconstructions (mean absolute error improves by 76.84%) and distributional similarity between real and synthetic data (maximum mean discrepancy improves by 13.85%). Our results highlight OmniDecVAEs potential as a lightweight model suitable for intelligent edge wearables and clinical healthcare, unifying processing requirements and abilities in a single model, through its enhanced representational capacity, modality-invariant spatial complexity (4.1M parameters), and real-time latency.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.