POSTER SESSION PROGRAM

September 1

Signal Decomposition and Estimation Methods

(R223-oral) Continuous-Time Signal Decomposition: An Implicit Neural Generalization of PCA and ICA
(R207-oral) Learning Rate Should Scale Inversely with High-Order Data Moments in High-Dimensional Online Independent Component Analysis
(R101-oral) LANM: Learned Atomic Norm Minimization for Superfast Gridless Spectral Compressed Sensing
(R18-oral) prNet: Efficient and Robust Phase Retrieval via Stochastic Refinement
(R200-oral) Coordinate Ascent Neural Kalman-MLE for State Estimation
(R33) Efficient algorithms for the Hadamard decomposition
(ISPS14) Outlier-Resilient Model Fitting via Percentile Losses: Methods for General and Convex Residuals
(R189) Efficient Algorithms for Estimating the Parameters of Mixed Linear Regression Models
(R6) Hankel Surrogate for Model-Bridge Simulation Calibration
(R75) Nonlinear Matrix Decomposition with the Sigmoid function
(R197) TSNPC: Learning from partially observed data using tensor network structured probabilistic circuits
(R180) Online Frequency Estimation with Adaptive Locally Differentially Private Mechanisms

NOCASA Data Challenge

Non-native Children's Automatic Speech Assessment Challenge (NOCASA)

Automated Pronunciation Scoring of Child L2 Learners with Score Calibration for Imbalanced Distributions

(oral) Comparison of End-to-end Speech Assessment Models for the NOCASA 2025 Challenge

Audio and Speech Processing

(R139-oral) DDL: A Dataset for Drone Detection and Localization from Multi-Channel Audio and a Deep Uncertainty-Aware Framework
(R203-oral) Audio Prototypical Network for Controllable Music Recommendation
(R206-oral) Re-Bottleneck: Latent Re-Structuring for Neural Audio Autoencoders
(R95-oral) Input Conditioned Layer Dropping in Speech Foundation Models
(R13) AudioMAE++: learning better masked audio representations with SwiGLU FFNs
(R214) HuBERT-Derived SSL Features and ECAPA-TDNN Matching for Robust Audio Deep‑Fake Detection
(R213) Efficient representation learning for music via likelihood factorisation of a variational autoencoder
(R198) Semi-Supervised Audio-Visual Action Recognition with Audio Source Localization Guided Mixup
(R128) SAGE: Spliced-Audio Generated Data for Enhancing Foundational Models in Low-Resource Arabic-English Code-Switched Speech Recognition
(R120) From Aesthetics to Human Preferences: Comparative Perspectives of Evaluating Text-to-Music Systems
(R2) Whisper speaker identification: leveraging pre-trained multilingual transformers for robust speaker embeddings
(R38) Prototypical contrastive learning for improved few shot audio classification
(R71) Tiny Noise-Robust Voice Activity Detector for Voice Assistants
(R90) Improving Phone Recognition through Informed Initialization and Path-Aligned CTC Loss
(R168) An analysis of associations between maternal vocalizations and infant stress recovery using speech emotion recognition models
(S33) Child speech assessment through large language model speech synthesis: Preliminary results

The Sampling-Assisted Pathloss Radio Map Prediction Challenge

Efficient Indoor Radio Map Prediction with Improved Transformers and Active Sampling Strategies
IRM-NET: An Enhanced Attention Networks for Indoor Radio Map Estimation
Sparse-Guided RadioUNet with Adaptive Sampling for the MLSP 2025 Sampling-Assisted Pathloss Radio Map Prediction Data Competition
Radio Map Prediction via Neural Networks with Ground Truth Shortcuts and Selective Sampling
U-Net Based Indoor Radio Map Prediction under Sparse Sampling
SAIPP-Net: A Sampling-Assisted Indoor Pathloss Prediction Method for Wireless Communication Systems
U-Net for Indoor Pathloss Prediction from Sparse Measurements with Physics-Informed Features
The Sampling-Assisted Pathloss Radio Map Prediction Competition

LLMs and Deep Learning Architectures

(R74) Convolutional spiking-based GRU cell for spatio-temporal data
(R68) Data Aware Differentiable Neural Architecture Search for Tiny Keyword Spotting Applications
(R46) An Alternating Algorithm for Neural Collapse in Deep Classifier Neural Network with Arbitrary Number of Classes
(R39) Neural-ANOVA: Analytical Model Decomposition using Automatic Integration
(R70) Asymptotic Study of In-context Learning with Random Transformers through Equivalent Models
(R85) Tackling Distribution Shift in LLM via KILO: Knowledge-Instructed Learning for Continual Adaptation
(R122) Post-inference guided transformer for anomaly interval localization in multivariate time series
(R177) Semantic Chunking and Chain-of-Thought Reasoning for RAG-based Document Processing

Computer Vision Applications

(R56) Efficient license plate recognition via pseudo-labeled supervision with Grounding DINO and YOLOv8
(R81) Pose-guided focal loss for enhancing vision transformers in continuous sign language recognition
(R8) Estimation of NDVI from UAV RGB imagery using deep learning models
(R138) Deep learning based polyculture structural phenotyping
(R146) 3D face morph generation using geometry-aware template inversion
(R147) Generating synthetic face recognition datasets using Brownian identity diffusion and a foundation model
(R155) Post-training quantization for Vision Mamba with k-scaled quantization and reparameterization
(R164) Attention and edge-aware band selection for efficient hyperspectral classification of burned vegetation
(R211) Bridging discrete and continuous: A multimodal strategy for complex emotion detection
(R210) MCFF-Det: Multispectral coarse-to-fine fusion for object detection
(R208) Efficient mouth alignment for visual speech recognition
(R171) Long range constraints for neural texture synthesis using sliced Wasserstein loss

September 2

ML for Neuroscience

(ISPS2) Latent representation learning for multimodal brain activity translation
(R42) UniPhyNet: A unified network for multimodal physiological raw signal classification
(R49) MicroWaveNet: Lightweight CBAM-augmented wavelet-attentive networks for robust EEG denoising
(R87) Towards generalizable learning models for EEG-based identification of pain perception
(R203) DECIFRA: Deep extraction of causally informed features via restricted architecture
(R178) Uncovering k-way balanced consensus communities in signed multilayer brain networks
(R170) Hypergraph overlapping community detection for brain networks

Learning Theory, Optimization and Algorithms

(R104-oral) Fast and Robust Training of Deep Learning Models with Multiplicative Adagrad
(R73-oral) Information Entropy-Based Scheduling for Communication-Efficient Decentralized Learning
(R80-oral) Model Recycling Framework for Multi-Source Data-Free Supervised Transfer Learning
(R162-oral) Meta-Tree: Bayesian Approach to Avoid Overfitting in Decision Trees and Analysis on the Application to Boosting
(R114) Model-Agnostic Uncertainty Calibration for Noisy Constraint Modeling in Bainitic Steel Optimization
(R173) Hess-MC²: Sequential Monte Carlo Squared using Hessian Information and Second Order Proposals
(ISPS5) Attention Augmented Structure-centric Bias Mitigation with Feature Disentanglement
(R136-oral) APA: Domain Generalization Using Frequency Based Augmentation
(R150-oral) Perturbation-based Multiview Graph Learning with Consensus Graph
(R3) Shapley-Based Data Valuation with Mutual Information: A Key to Modified K-Nearest Neighbors
(R72) Toward Sustainable Continual Learning: Detection and Knowledge Repurposing for Reoccurring Tasks
(R105) Online Topology Identification of Higher-Order Cell Structures
(R215) Benefits of Online Tilted Empirical Risk Minimization: A Case Study of Outlier Detection and Robust Regression
(R184) EMORF-II: Adaptive EM-Based Outlier-Robust Filtering with Correlated Measurement Noise

Special Session, Applications of AI in the Analysis of Cultural and Artistic Heritage

(S26-oral) Solving Jigsaw Puzzles in the Wild: Human-Guided Reconstruction of Cultural Heritage Fragments
(S20) The Cow of Rembrandt - Analyzing Artistic Prompt Interpretation in Text-to-Image Models
(S11) Speaking Images: A Novel Framework for the Automated Self-Description of Artworks
(S6) Experimenting Active and Sequential Learning in a Medieval Music Manuscript
(S25) Deep Learning for Fine-Grained Classification of Montelupo Majolica: Benchmarking and Explainability
(S30) DFA-CON: A Contrastive Learning Approach for Detecting Copyright Infringement in DeepFake Art
(S31) Restoration and Enhancement of Historical Manuscript Images Using Diffusion Model
(S10) Multimodal Artwork Topic Modeling via Fine-Tuned CLIP and Knowledge-Driven Prompts

Special Session, LEAP: Low-Energy AI For Edge Learning and Processing

(S32-oral) Improving Communication-Efficiency for Decentralized Federated Clustering
(S1) Unseen Speaker and Language Adaptation for Lightweight Text-To-Speech with Adapters
(S2) Utilizing Dynamic Sparsity on Pretrained DETR
(S7) Neural Successive Cancellation Decoder for Polar Codes Using Analog In-Memory Computing with Memristors
(S19) Self-Supervised Learning at the Edge: The Cost of Labeling
(S21) Energy-Information Trade-Off in Self-Directed Channel Memristors

Special Session, Large Vision Language Models (LVLMs) and their Application to Document Understanding

(S22-oral) Trust the Model: Compact VLMs as In-Context Judges for Image-Text Data Quality
(S4) Multi-Agent Interactive Question Generation Framework for Long Document Understanding
(S14) Enhanced Arabic Text Retrieval with Attentive Relevance Scoring
(S28) Learning or Cheating? Assessing Data Contamination in Large Vision-Language Models
(S29) ColFlor: Towards BERT-Size Vision-Language Document Retrieval Models

Applications in Wireless, Radio and Radar Communications

(R11) Few-Shot Radar Signal Recognition through Self-Supervised Learning and Radio Frequency Domain Adaptation
(ISPS9) Signal Processing Challenges in Automotive Radar
(ISPS8) Advancing Single-Snapshot DOA Estimation with Siamese Neural Networks for Sparse Linear Arrays
(ISPS7) Advancing High-Resolution and Efficient Automotive Radar Imaging Through Domain-Informed 1D Deep Learning
(R50-oral) RadioTrace: Bridging Diffusion Priors and RSS Measurements for Accurate Radio Map Estimation
(R62) Robust Imputation SwinLSTM for Spectrum Map Prediction of Incomplete Data
(R127) Online Gaussian Process for Dynamic Radio Map Updating
(R10) Signal Prediction for Loss Mitigation in Tactile Internet: A Leader-Follower Game-Theoretic Approach
(R154) Benchmarking Transfer Learning in Passive Sonar: An Evaluation Study
(R219) Robust and Efficient Kernel-Based Digital Self-Interference Cancellation Using A Priori Knowledge in Full-Duplex Transceivers

Diffusion, Generative and Representation Learning

(R37) MPRDiff: Mixed Precision Restorative Diffusion Model with Incoherence Processing
(R14) Backdoor Inversion in Neural-Activation Space
(R22) Contrastive Disentanglement Learning for Empathetic Generation
(R51) Compression Beyond Pixels: Semantic Compression with Multimodal Foundation Models
(R21) Diffusion-Based Connectionist Temporal Classification
(R5) Foundation Model-Aided Video Semantic Communication: Framework Design and Prototype Validation
(R4) FreNBRDF: A Frequency-Rectified Neural Material Representation
(R95) Input Conditioned Layer Dropping in Speech Foundation Models
(R204-oral) Ptychographic Image Reconstruction from Limited Data via Score-Based Diffusion Models with Physics-Guidance
(R58) Stability and Performance Analysis of Diffusion Learning for Two-Network Competing Problems

September 3

ML for Medical Applications

(R137-oral) Cycle-Consistent Diffusion Model with Vessel-Aware Attention for Endoscopic Image Translation
(ISPS3) DiffKillR: Killing and Recreating Diffeomorphisms for Cell Annotation in Dense Microscopy Images
(ISPS1) ImageFlowNet: Forecasting Multiscale Image-Level Trajectories of Disease Progression with Irregularly-Sampled Longitudinal Medical Images
(R24) SVD Based Least Squares For X-Ray Pneumonia Classification Using Deep Features
(R31-oral) RSR-NF: Neural Field Regularization by Static Restoration Priors for Computed Dynamic Imaging
(R182-oral) Closing the Gap in Multimodal Medical Representation Alignment
(R7) Adaptable Non-parametric Approach for Speech-based Symptom Assessment: Isolating Private Medical Data in a Retrieval Datastore
(R15) "Digital Washing" of Semen Time-lapse Images
(R148) StrokeVision-Bench: A Multimodal Video and 2D Pose Benchmark for Tracking Stroke Recovery
(R163) Does Language Matter for Early Detection of Parkinson's Disease from Speech?

Human Activity Recognition, Physiological Signals & Wearables

(ISPS13) OSR: Toward Developing Efficient Federated Learning-based Human Activity Recognition using Optimal Server Representations
(R41) Compressed and Lightweight CNN for Real-Time Parkinson's Tremor Detection from Wearable IMU Data
(R142) Subject Invariant Contrastive Learning for Human Activity Recognition
(R98) Graph Structure Learning with Local Connectivity Refinement for Improved Physiological Emotion Recognition
(R188) GEMS: Group Emotion Profiling through Multimodal Situational Understanding
(ISPS11) Interbeat Interval Filtering
(R20) Classification Filtering
(R27) Embedded Inter-Subject Variability in Adversarial Learning for Inertial Sensor-Based Human Activity Recognition

Decision Making, Bandits, and Recommendation Systems

(ISPS4) A Distillation-based Future-aware Graph Neural Network for Stock Trend Prediction
(R156) Learning from Multiple Noisily Optimal Demonstrators in Stochastic Multi-Armed Bandits
(R194) Networked Contextual Bandits with Anomaly-Aware Learning
(R193) State Prediction for Offline Reinforcement Learning via Sequence-to-Sequence Modeling
(R190) Poisson-based Modeling and Curvature-aware Optimization for Neural Collaborative Filtering in Recommendation Systems
(R157) Integrating Adaptive Prediction with an Optimization-based Methodology for Data-Driven Efficiency Evaluation in Education
(ISPS6) What Does an Audio Deepfake Detector Focus on? A Study in the Time Domain
(R141) CollabPersona: A Framework for Collaborative Decision Analysis in Persona Driven LLM-based Multi-Agent Systems

Federated & Decentralized Learning

(ISPS10) Federated Domain Generalization with Label Smoothing and Balanced Decentralized Training
(ISPS12) Enhancing Federated Learning Convergence With Dynamic Data Queue and Data-Entropy-Driven Participant Selection
(R222) Provable Reduction in Communication Rounds for Non-Smooth Convex Federated Learning
(R111) Joint Graph Estimation and Signal Restoration for Robust Federated Learning
(R107) Memory-Efficient Correlated Noise for Locally Differentially Private Momentum in Distributed Learning
(R65) DeMem: Privacy-Enhanced Robust Adversarial Learning via De-Memorization

Special Session, Decoding the Brain Time Series

(S12-oral) Toward a Gaze-Independent Brain-Computer Interface Using the Code-Modulated Visual Evoked Potentials
(S3S) Assessing the Capabilities of Large Brainwave Foundation Models
(S5) Uncertainty Quantification for Motor Imagery BCI - Machine Learning vs. Deep Learning
(S8) On the Role of Low-Level Visual Features in EEG-Based Image Reconstruction
(S9) Deep Learning of Mesoscale Cortical Dynamics for Real-Time Classification of Forelimb Movement in Mice
(S16) Interpretability of Riemannian Tools Used in Brain-Computer Interfaces
(S18) AbsoluteNet: A Deep Learning Neural Network to Classify Cerebral Hemodynamic Responses of Auditory Processing
(S23) Riemannian Fusions of EEG-Based Features for Motor Imagery Detection under Propofol Sedation
(S27) Evaluating Manifold Alignment of Motor Imagery for Transfer Learning in EEG-Based BCIs
(S9) Deep Learning of Mesoscale Cortical Dynamics for Real-Time Classification of Forelimb Movement in Mice

Data Competition, VEELA - Vessel Extraction and Extrication for Liver Analysis

Hepatic Vessel Segmentation and Classification in CTA Images Using NNU-Net with Centerline Regression
Hybrid Boundary Sensitive Tversky 3D U-Net for Liver Vessel Segmentation
Veela Challenge - Vessel Extraction and Extrication for Liver Analysis