Sufficiently Reduced Distributional Regression (SRDR) is a generative approach that merges conditional distribution estimation with nonlinear sufficient dimension reduction (SDR). By framing SDR as a risk minimization problem using strictly proper scoring rules, SRDR jointly learns a dimension reduction map and a generative prediction model through minimization of the energy score, which can be estimated via sampling. The method extends to multi‑environment data and classification, and theoretical results show convergence of estimated conditional distributions in energy distance, implying asymptotic sufficiency. In experiments on CT slice localization, superconductivity data, and digit classification, SRDR recovers low‑dimensional sufficient structure and matches or surpasses state‑of‑the‑art nonlinear SDR methods in representation quality and predictive performance.
By Alexander Henzi, Tiange Liu, Xinwei Shen
The paper introduces Flow-Based Distribution Matching (FBDM), a non‑adversarial framework that learns self‑supervised representations using explicit geometric references and spherical conditional velocity regression. FBDM assigns augmented image views to shared target references while limiting reference usage, and employs an alignment loss to bring view representations closer. Experiments on datasets from CIFAR to ImageNet demonstrate that FBDM performs nearly as well as adversarial DM, outperforms existing SSL methods, and achieves a 1.48‑ to 1.83‑fold speedup with minimal GPU memory increase, while a theoretical analysis bounds downstream misclassification rates in terms of the pretraining loss.
By Yuling Jiao, Wensen Ma, Houduo Qi, Defeng Sun
RAPTOR is a novel time‑domain simulation framework for power systems that uses a physics‑informed random‑projection neural network built from fixed Gaussian radial basis functions to represent the trajectory of hybrid differential‑algebraic equations over a time interval. By combining these basis functions with a nonlinear solver such as Newton‑Raphson, RAPTOR can capture complex multi‑timescale dynamics over longer intervals, reducing the number of simulation advances and nonlinear iterations needed. Numerical tests on RMS and EMT benchmarks demonstrate that RAPTOR achieves accurate solutions with significantly fewer steps and can outperform traditional methods like Radau and the trapezoidal rule, achieving speedups of over tenfold in some cases.
By Petros Ellinas, Benjamin Vilmann, Spyros Chatzivasileiadis, Johanna Vorwerk
The paper introduces CLASP‑Ar, a cloze‑style prompting method for Arabic stance detection that replaces complex multitask learning and ensembles with a single masked language modeling prompt. By combining the target, predicted sentiment, and text into one prompt and constraining the [MASK] prediction to a verbalizer‑defined label set, the approach aims to simplify the task while maintaining performance.
By Bhuvanesh Verma, Ali Abusaleh, Alexander Mehler
The paper introduces Higher-Order Modular Attention (HOMA), a new attention mechanism that combines standard pairwise self‑attention with an explicit triadic attention pathway. HOMA uses overlapping blocks, local windows, and a low‑rank projection to make triadic interactions tractable. Experiments on controlled PARITY and MATCH3 tasks, as well as TAPE benchmarks, show that HOMA matches or outperforms matched pairwise and purely triadic baselines, especially when dependencies extend beyond triadic order, and it often converges faster and uses parameters more efficiently.
By Shirin Amiraslani, Xin Gao
MLPerf Automotive is the first standardized public performance benchmark for evaluating machine learning systems used in automotive AI acceleration. Developed by MLCommons, it addresses the unique constraints of automotive workloads—such as sensor suites, safety, and real‑time processing—that existing benchmarks cannot handle. The benchmark covers perception tasks (2D/3D object detection, semantic segmentation), end‑to‑end driving, and infotainment, and includes curated models, methodology, and reference implementations available on GitHub.
By Radoyeh Shojaei, Predrag Djurdjevic, Mostafa El-Khamy, James Goel, Kasper Mecklenburg, P{\i}nar Muyan-\"Oz\c{c}elik, John Owens, Tom St. John, Jinho Suh, Arjun Suresh
GOMA (Graph-Optimized Multimodal Alignment) introduces a dual-embedding approach for multimodal retrieval, separating content embeddings supervised for paired identity from semantic embeddings trained with cross-modal pairs and observed relationships. The method fuses these embeddings, applies semantic agreement to weight graph edges, and uses restart graph propagation to reinforce the initial signal, enabling both single-modality and dual-attribute retrieval. Across six datasets and four tasks, GOMA outperforms 14 external methods on 14 primary metrics, with controlled experiments highlighting the impact of separate supervision, graph regularization, and semantic-guided propagation.
By Xu Wang, Xunkai Li, Yinlin Zhu, Rong-Hua Li, Guoren Wang
Time‑Varying Bayesian Optimization (TVBO) is a framework for optimizing expensive, noisy black‑box functions that change over time. Existing TVBO algorithms assume observations are taken at a constant frequency, an assumption that becomes unrealistic as Gaussian‑process inference scales with the cube of dataset size. This paper relaxes that assumption, derives the first upper regret bound that incorporates variable sampling frequency, and uses the analysis to give practical guidance on dataset sizes and stale‑data policies. An algorithm (BOLT) that follows these recommendations outperforms the current state‑of‑the‑art TVBO methods on both synthetic and real‑world experiments.
By Anthony Bardou, Patrick Thiran
BanglaTurn is a new corpus of 35,374 Bangla podcast speech samples, each 3 to 15 seconds long, labeled for end‑of‑turn detection through speaker diarization, an LLM pass, and human verification. A Whisper‑based model with task‑specific classification heads achieves 84.33 % accuracy on a balanced test set, outperforming the Smart‑Turn v3 baseline (69.28 %) and reducing the false‑negative rate from 51.57 % to 7.55 %, though with a higher false‑positive rate. The study also details the contributions of encoder‑layer fine‑tuning, multi‑scale pooling, INT8 quantization, and reports inference latency of 165–191 ms on CPU.
By Mizbaul Haque Maruf
The paper introduces Heartian, a physiology‑aware framework that augments Gaussian head avatars with cardiac‑cycle‑dependent albedo modulation, enabling the encoding of remote photoplethysmography (rPPG) signals. By supervising with synchronized contact PPG, the method models the cardiac waveform as a sum of two Gaussians and learns per‑frame spatial residuals via a lightweight MLP. Experiments on 152 stationary recordings from UBFC‑rPPG, PURE, and MMPD show heart‑rate estimation errors as low as 0.29 bpm MAE and 0.38 % MAPE, while preserving reconstruction quality with negligible PSNR loss.
By Xiaoyue Fan, Jose Echevarria, Akshay Paruchuri, Kaan Ak\c{s}it
STRAND is a new benchmark that tests multimodal large language models’ ability to track objects, their states, and relationships over time in videos. It evaluates intermediate reasoning by breaking queries into sub‑questions and uses Faithful Accuracy to ensure all parts of an answer are correct. The authors also propose an object‑centric framework that builds structured trajectories and shows reduced hallucinations and better temporal consistency compared to existing models.
By Thong Nguyen, Tri Cao, Khoi Le, Cong-Duy Nguyen, Quynh Vo, See-Kiong Ng, Bryan Hooi Kuen-Yew
Token Clustering and Semantic Sequence Mamba (STMamba) is a new approach for hyperspectral image classification that organizes sparse tokens into semantically coherent sequences. It uses a hierarchical encoder-decoder with a Token Clustering Module (TCM) to select semantic tokens and a Cross-scale Neighborhood Attention (CNA) Upsampler to restore dense features. At the micro level, density-aware clustering and a quadtree-based dynamic selection keep sparse, spatially distributed tokens, while Spatial and Spectral Semantic-wise Sequencing Mamba (SWSM) modules capture long-range spatial and spectral dependencies within homogeneous semantic token sequences. Experiments on three large-scale benchmark datasets show that STMamba outperforms state‑of‑the‑art methods in both quantitative and qualitative metrics.
By Yimin Zhu, Mahmood Elahi, Lincoln Linlin Xu
PePESeg3D introduces perception priors into a multi‑scale 3D Gaussian segmentation pipeline, integrating monocular depth and mask constraints during geometry reconstruction and dense depth‑color cues with view‑consistent centroid supervision during contrastive feature learning. This dual‑stage approach aligns geometry with semantic structure and compensates for incomplete mask supervision from 2D foundation models. Experiments on SPIn‑NeRF, LERF‑Mask, and NVOS benchmarks show state‑of‑the‑art performance in both multi‑scale segmentation and scene reconstruction.
By Sungjae Choi, Seunghee Koh, Junmo Kim
PlenoCI introduces a novel feature derived from the plenoptic field of radiance representations like 3D Gaussian Splatting, capturing rich visual behaviors while ignoring Lambertian textures. By computing closed‑form analytic plenoptic derivatives, the method robustly detects 5D structures and reduces false positives by two orders of magnitude compared to prior work. PlenoCI is applied to change classification, achieving a 25.7% mIoU improvement on CL‑Splats and a balanced accuracy of 0.735 for distinguishing geometric from appearance‑based changes.
By Jason Lai, Chamuditha Jayanga Galappaththige, Niko Suenderhauf, Dimity Miller, Donald G. Dansereau
MoVISA introduces Multi-Token Reasoning for Video Object Segmentation, using multiple segmentation tokens (e.g., SEG0, SEG1) instead of a single token to better localize multiple objects over time. This approach enhances fine-grained alignment between language prompts and spatio-temporal mask predictions, leading to improved performance and interpretability. On benchmarks such as MeViS, DAVIS17, ReVOS, and Ref-Youtube-VOS, MoVISA achieves significant gains, notably a 13.2% J and F improvement on MeViS and an 8.4% J and F improvement on ReVOS.
By Ruining Zhao, Ho Kei Cheng, Alexander G Schwing
The SEE Challenge 2026 invites participants to restore RGB images using synchronized event camera data and a target brightness statistic across a wide illumination range. Using the SEE-600K dataset of 610,126 image‑event pairs from 202 real‑world scenes, teams compete under an open‑system protocol, with PSNR as the primary ranking metric and SSIM as a secondary measure. Fifteen valid submissions were evaluated, revealing closely spaced top scores and consistent local errors under severe underexposure, while the report also examines exposure subsets, semantic test cases, shared failure patterns, and system design choices.
By Yunfan Lu, Mingchao Xu, Hanyu Zhou, Shaoyu Liu, Haoyue Liu, Peiqi Duan, Shihan Peng, Yinqiang Zheng, Boxin Shi, Gim Hee Lee, Hui Xiong, Davide Scaramuzza
The paper introduces Anomaly‑LR, a defect‑grounded latent reasoning framework for industrial anomaly detection that builds a global understanding of an image and then refines anomaly‑relevant representations directly in visual latent space. It also presents IAD‑LR‑22K, a new instruction dataset with 22,228 image‑question pairs and detailed annotations. Experiments demonstrate that Anomaly‑LR outperforms comparable‑scale methods on multiple IAD benchmarks without needing external references or tools.
By Jaron Yeh, Yen-Wei Chang, Jiang Liu, Shao-Yuan Lo
SpectralCTGaussians introduces a novel spectral CT reconstruction and basis material decomposition technique that employs 3D Gaussian splatting with per‑Gaussian material fractions and energy‑dependent basis functions. By jointly optimizing these parameters across all energy channels via a differentiable polychromatic forward model, the method achieves superior novel view synthesis and higher PSNR for spectral CT volume reconstruction compared to traditional and learning‑based baselines. It also provides one‑step material decomposition with direct RGB segmentation and recovers the photoelectric basis more accurately than existing pipelines.
By Reinout Vos, Saptarshi Neil Sinha, Michael Weinmann
TopoFuse introduces a topology-aware tri-planar fusion method for 3D cryo-electron tomography segmentation. It replaces traditional loss penalties with a differentiable projection operator that identifies and sparsely edits critical voxels to enforce specified topological constraints. The approach achieves a 54% reduction in Betti number error, a 4.6-point Dice improvement, and edits only 3.1% of voxels across three benchmarks.
By Rohit Kumar Salla, Neelesh Gupta, Xingjian Li, Min Xu
EndoFSA is a GAN-based model designed for endoscopic few-shot image generation, addressing the scarcity of pathological samples in wireless capsule endoscopy (WCE) data. It adapts a generator pretrained on abundant normal images to abnormal domains by updating only a small set of rank-constrained modulation parameters while keeping the rest of the weights frozen, thereby preserving anatomical priors and preventing mode collapse. The method incorporates perceptual boundary regularization and cluster-wise diversity control, operates without pixel-level annotations, and demonstrates that synthetic abnormal images can match real images in downstream classification performance.
By Panagiota Gatoula, Grigoris Karypidis, Dimitris K. Iakovidis