Shravan Venkatraman, Wenshuai Zhao, Mohammad Hassan Vali +1cs.CV
We introduce S$^3$T (Self-Supervised Self-Distillation over Time), which, to the best of our knowledge, is the first fully self-contained framework for continuous video state tracking. Our method treats temporal sampling density as privileged information, based on the hypothesis that a denser view of the same clip recovers the running state more accurately. This view serves as the teacher, while a sparse-view student with the same weights learns to match its next-token distribution. The model generates its own target, so training requires no labels, separate teacher, or reward signal, and adds no inference cost. On LLaVA-OneVision-2-8B, S$^3$T improves VSTAT accuracy by $+1.74$ as a single model, $+2.38$ with souping, and $+2.70$ with additional vision-encoder adaptation, while prior self-evolving methods leave state tracking largely unchanged. The capability learned from unlabeled synthetic clips transfers to real videos, improving performance by $+7.95$ on VSTAT-YouTube state-tracking questions and $+4.50$ on MVBench Action Count.
Advances in neural rendering have enabled high-fidelity multi-view reconstruction of 3D scenes. However, free-form non-rigid shape editing remains a significant challenge. Point-based neural representations are highly desirable for multi-view reconstruction because they lack fixed connectivity, which does not constrain the learned surface topology to that of the initialization. Yet this same property causes point-based representations to struggle with holes and surface discontinuities under large deformations. To address this, we propose a novel self-supervised method to enable point-based representations to adapt to large deformations without requiring ground truth multi-view images of deformed geometry. The key idea is to generate random deformations and to ensure consistency in the predicted surface before and after deformation. In particular, the surface prediction from the deformed point cloud should be the same as the deformation applied to the surface prediction from the original point cloud. We incorporate our approach into attention-based point representations, which differ from splatting-based point representations in their use of a learned interpolation kernel between points as opposed to a Gaussian kernel around each point. This learned interpolation kernel can learn to adapt to large deformations, without requiring addition or removal of points. We show that our framework significantly enhances its robustness to large deformations. Experiments on synthetic geometry editing benchmarks (Neural Editor, Objaverse) demonstrate that our approach outperforms existing point-based methods in zero-shot editing and significantly reduces artifacts. Furthermore, qualitative results on the DTU and Mip-NeRF 360 datasets demonstrate our method's effectiveness on real-world scenes.
Lucas Gerken Starepravo, Henry Broadley, Steven Lind +1physics.comp-ph cs.LG
Meshfree methods such as smoothed particle hydrodynamics (SPH) with kernel corrections, radial basis function-generated finite differences (RBF-FD), and the local anisotropic basis function method (LABFM) construct discrete differential operators by imposing polynomial consistency on a local stencil. For stencils containing more nodes than there are consistency constraints, the resulting linear system is underdetermined, and the remaining degrees of freedom are fixed implicitly by the choice of kernel, basis preconditioning, or a minimum-norm condition. Polynomial consistency constrains the operator only in the low-wavenumber limit, and no part of the construction selects for accuracy at the wavenumbers where fine-scale content resides. We introduce Spectral-like Neural Discretisation (SpeND), in which the choice of those degrees of freedom is cast as a learning problem: stencil weights are parametrised by a neural network conditioned on the local node geometry, trained to approximate the modal response of a spectral operator over the resolvable band. A hard-constrained projection layer maps the network output onto the affine subspace of consistent weights, so that polynomial consistency holds exactly by construction rather than as a penalty. Training is self-supervised and physics-agnostic, requiring no reference solutions; the objective minimises dispersion and dissipation error over a prescribed band-limited function space. Modal analysis on disordered two-dimensional node distributions shows that the learned fourth-order operator follows the exact response over a substantially wider band than either explicit LABFM at equal stencil size or fourth-order finite differences on a structured grid, whilst recovering the expected fourth-order convergence rate under refinement.
High-resolution Synthetic Aperture Radar (SAR) imagery is critical for precision analysis such as automatic target recognition, yet its acquisition is costly. Although generative image super-resolution (ISR) models offer a promising alternative, current smooth-approximation based diffusion frameworks often struggle to preserve the coherent scattering statistics, causing stochastic structural distortions that are less consistent with real SAR physics. To address this, we propose Semantic Prototype-Guided Super-Resolution (ProSR), reformulating SAR ISR as a semantically-guided discrete token prediction task within a quantized latent space. By mapping signal features to discrete scattering primitives, ProSR preserves the impulsive nature of SAR without over-smoothing. Furthermore, we integrate a Self-Supervised Learning backbone into SAR ISR to extract label-free semantic priors, overcoming label scarcity. Guided by these priors, we introduce Semantic-Aligned Detail Encoding to decouple high-frequency signals into discrete scattering primitives. In parallel, the Semantic Prototype Map Generator explicitly constructs semantic prototype maps, allowing Prototype-Map-Guided Attention to route the information flows within identical categories and mitigate inter-class interference. To validate our approach, we present a large-scale 0.25m resolution benchmark from the Umbra Open Dataset. Experimental results show ProSR achieves superior visual quality while preserving essential scattering characteristics required for practical SAR applications.
Diffusion MRI requires repeated k-space acquisitions over multiple diffusion-encoding directions, making acquisition time dependent on both spatial and angular sampling. Existing joint k-q methods either associate directional parameters with fixed voxels or separate spatial reconstruction from angular completion. However, diffusion-weighted images acquired under different directions share the same anatomical organization, while their local signal intensities vary with diffusion encoding. Existing formulations do not fully exploit the complementarity between shared anatomy and direction-dependent signal variation. Consequently, residual spatial errors may be misinterpreted as genuine angular variation and propagated to unobserved directions. We propose a subject-specific spatial-angular Gaussian field for self-supervised joint k-q dMRI reconstruction. Shared 3D Gaussian primitives provide local spatial support, with each primitive carrying a continuous q-conditioned tensor-residual response. The signal at each location is synthesized from multiple overlapping primitive responses, coupling neighboring spatial regions and diffusion directions. The field is progressively optimized from undersampled k-space measurements of observed directions, without fully sampled targets or held-out-direction supervision. Experiments on three HCP diffusion shells under multiple acceleration settings demonstrated consistent improvements in missing-direction DWI reconstruction, tensor-derived metrics, and principal diffusion orientation estimation.
Jialin Liu, Zhaorui Zhang, Ray C. C. Cheungcs.IR cs.AI
Multimodal Recommender Systems (MRSs) typically rely on a flawed "modality harmony" assumption, presuming that multimodal features are inherently beneficial and strictly aligned with users' collaborative interaction patterns. However, modality-topology conflicts are ubiquitous in real-world scenarios due to deceptive visual clickbaits and mismatched semantics. Blindly integrating these noisy modalities inevitably pollutes the pristine collaborative space, causing severe representation distortion. To address this, we propose Orthogonal purification and topology-guided MoE for conflict-aware multimodal Recommendation (OrthoRec). At its core, OrthoRec introduces Collaborative-Guided Orthogonal Purification (CGOP), which geometrically decouples multimodal features into directions parallel and orthogonal to a pure collaborative anchor. By adaptively truncating the orthogonal noise with an energy-preserving normalization, CGOP rectifies deceptive semantic directions while preserving the modality's intrinsic representation capacity. Furthermore, we design a Topology-Aware Routing Mixture-of-Experts (TAR-MoE). Guided by the collaborative topology, TAR-MoE employs decoupled sigmoid gating to break the zero-sum bottleneck of traditional softmax attention, autonomously determining the injection scale for each purified modality. Finally, a safe-SSL objective is introduced to dynamically penalize the forced contrastive alignment of contradictory pairs. Experiments on three real-world Amazon datasets show that OrthoRec consistently outperforms competitive recent baselines and exhibits improved robustness under modality noise and item sparsity.
In computational pathology (CPath), developing omni-modal self-supervised learning (SSL) models that integrate histology, genomics, and clinical reports enables transferable representation learning for whole slide images (WSIs). Existing approaches implicitly force heterogeneous modalities into a uniform latent space by contrastive alignment, causing modality collapse where unique, synergistic diagnostic signals (termed as $\mathrmΦ$) are discarded in favor of trivial redundancy. We hypothesize that the strongest task-agnostic SSL training signal stems from distilling the synergistic interactions over merely aligning shared redundancy. To this end, we introduce \textsc{$\mathrmΦ$-Omni}, a synergistic information disentanglement framework grounded in Partial Information Decomposition (PID) theory for slide representation learning. Unlike standard contrastive approaches, \textsc{$\mathrmΦ$-Omni} employs a Synergistic Information Bottleneck (SIB) regulated by the proposed $\mathrmΦ\text{ID}$ objective, which explicitly suppresses marginal redundancy while maximizing irreducible synergy, thereby distilling high-order cross-modal interactions. Following pretraining on breast ($n$=1031) and lung ($n$=919) cohorts, \textsc{$\mathrmΦ$-Omni} demonstrates superior few-shot performance across five independent external datasets spanning eight tasks compared to supervised and SSL baselines. Source code is available here.
Vision Transformers (ViT) excel in semantic understanding but fail to discriminate between object instances (e.g., identical embeddings for two dogs), limiting their use in instance-level tasks such as object detection and instance segmentation. We propose Contrastive Vision Transformer (CoViT), a self-supervised learning framework that injects instance-awareness into ViT through geometry-guided contrastive learning. CoViT uniquely coordinates ViT's attention maps and embeddings by constructing triplets: (1) Attention-guided masking: Refine multi-head attention via adaptive thresholding and morphological operations to generate instance masks, identifying foreground anchors; (2) Hardest contrastive mining: For each anchor, computing pairwise embedding similarities to select the intra-instance hardest positive (least similar patch within its mask) and inter-instance hardest negative (most similar patch from other instances), with intra-instance regions masked during negative search. These triplets drive a contrastive loss that simultaneously compresses intra-instance variance and expands inter-instance margins, forcing ViT to discern subtle geometric and appearance differences between instances. CoViT consistently achieves stable performance gains of over 2 AP points across multiple instance-level perception tasks by using ViT as backbone architecture. Notably, CoViT requires no extra decoders or labels, demonstrating that a pure ViT can learn instance-aware representations via inherent attention priors and targeted contrastive constraints. Code and models will be released.
Thibaut Loiseau, Guillaume Bourmaud, Vincent Lepetitcs.CV
Self-supervised pre-training via cross-view completion learns strong features for 3D vision from co-visible regions of image pairs. However, the reference view provides little information for reconstructing non-co-visible patches, implicitly yielding a monocular training signal in these regions. We introduce Gekko, which turns this limitation into a useful signal. The relative improvement of the cross-view reconstruction error over a masked-autoencoder error is a self-supervised proxy for co-visibility: large improvements indicate co-visible regions, negligible ones non-co-visible areas. Gekko is a network, trained from scratch, that jointly performs cross-view completion, masked autoencoding, and per-pixel prediction of this relative improvement, providing an additional binocular signal for all masked regions without any ground-truth 3D annotation. Under identical architectures and training data, Gekko consistently outperforms CroCo on zero-shot correspondence estimation, relative pose estimation, and pointmap regression, with up to 6 times higher accuracy at the strictest relative-pose threshold and a 22% drop in end-point error on ETH3D. The extra channel it learns is itself a strong co-visibility detector on unseen scenes, and Gekko's frozen features outperform released cross-view backbones of comparable or larger size. It can also be trained directly from raw videos with a simple stride-based curriculum, removing the cumbersome 3D preprocessing prior methods require while matching models trained on curated data. Code and pre-trained models are publicly available.
Lucas Cunha, Lucas Sotomaior, Lucas Gasperin +3cs.CV
Face forgery detectors often achieve strong results on controlled benchmarks, but their reliability under realistic image degradations remains limited. This paper presents a standardized benchmark for face forgery detection using the Multi-Dimensional Face Forgery Image (MFFI) dataset and evaluates performance on both clean and degraded test partitions. We compare six model families, including convolutional networks, transformer-based models, and a frozen self-supervised DINOv3 backbone, across spatial, spectral, and hybrid input representations. The results show that clean-set performance is not a reliable indicator of robustness under compression, resizing, and blurring. Xception with RGB obtains the best clean performance, reaching 0.884 mean ROC-AUC, but degrades substantially on the harder partition. In contrast, frozen DINOv3 achieves the strongest degraded-set result, with 0.726 mean ROC-AUC, while training only a linear classification head. The representation analysis indicates that Fourier-domain cues are most useful when combined with RGB information, whereas purely spectral inputs consistently underperform spatial representations. Qualitative attribution maps further suggest that convolutional detectors focus on localized artifacts, while DINOv3 relies on broader facial structure. These findings reinforce the need for degraded evaluation protocols and highlight self-supervised visual representations as a promising direction for robust face forgery detection. Our source code is publicly available at https://github.com/lucasdocunha/FaceForgery-Benchmark/.
Dense self-supervised learning (SSL) is a powerful paradigm for learning without annotations the local descriptors required to solve dense medical imaging tasks. We present Pix2Rep-v2, a framework for SSL of pixel- and voxel-level representations suitable for few-shot downstream applications. Pix2Rep-v2 addresses the main challenges of dense SSL by leveraging a redundancy reduction objective at the pixel-level with a principle of equivariance of dense representations, that scales efficiently to 3D or wide field-of-view applications. We evaluate our method on four datasets, across multiple tasks, multiple modalities and anatomical structures using multiple backbones in 2D and 3D, and under various data regimes. As an alternative to linear probing or full fine-tuning on the downstream task, we also propose an in-context variant, without downstream training, based on a dense prototype approach. Pix2Rep-v2 shows substantially higher data-efficiency in few-shot scenarios compared to fully supervised baselines, and is competitive with the state-of-the-art e.g., +9.3 Dice points in one-shot segmentation on the M&Ms-2 dataset. Our code and pre-trained models are publicly available at https://github.com/BioMedTP/pix2rep-v2.
Athira J. Jacob, Puneet Sharma, Daniel Rueckertcs.CV
Cardiac magnetic resonance (CMR) imaging provides complementary information on cardiac anatomy, function, and tissue characterization across multiple sequences and views. In this work, we investigate foundation model pretraining for 2D CMR and introduce CMRVision, a CMR-specific foundation model trained using DINOv3-style self-supervised learning on a multi-center, multi-sequence cohort of 36 million CMR images. We systematically evaluate architectural and training design choices for domain-specific pretraining. CMRVision is evaluated on two downstream tasks: multi-task segmentation across cine, late gadolinium enhancement (LGE), and mapping sequences, and cine view classification. Our experiments show that CMR-specific pretraining, smaller patch sizes, and patch-level objectives consistently improve downstream performance. Across a multi-task segmentation benchmark, CMRVision achieved the strongest overall performance, outperforming prior natural-image (NI), medical-image, supervised, and CMR foundation model baselines. Improvements were modest but consistent across structures and sequences, with Dice scores ranging from 0.940-0.967 for LV and 0.855-0.905 for myocardium, and reaching 0.929 for RV, 0.920 for LA, and 0.931 for RA. The largest gains were observed for myocardium segmentation in LGE and mapping images. In a zero-shot segmentation task on unseen LGE long-axis views, the model achieved an average Dice score of 0.692, demonstrating cross-view generalization. For cine view classification, CMRVision achieved the highest average accuracy (0.906), compared to prior methods reported in the literature. These results highlight the potential of CMRVision to support robust and generalizable cardiac MRI analysis across multiple sequences and views.
Tabular deep learning (TDL) leverages neural networks (NN) to extract patterns from tabular data. Traditional TDL methods follow a supervised learning paradigm, where a target feature is explicitly given. In this work, however, we explore a different approach by employing deep NNs to learn relationships among individual columns within a given table. We investigate whether NNs can predict the values of arbitrarily selected columns in a given table based on the remaining known columns. We call this problem In-Table Prediction (ITB), which is slightly different from table imputation methods and the pretraining task of TDL. Three potential usage scenarios are identified, which, to our best knowledge, have not been extensively studied in the literature. A self-supervised learning approach is applied to address this problem by randomly selecting columns to be masked out and used as learning targets. This work focuses on tabular datasets containing only continuous features. To handle missing values in continuous features, a novel neural layer is proposed to embed both numerical and empty values. Synthetic data is generated based on predefined column relationships, with empty values inserted using two distinct mechanisms. Additionally, an adapted masking strategy is employed to create test data. Performances of three NN architectures, namely MLP, Resnet and Transformer, are evaluated using the generated synthetic data. We conclude that, the attention-based structure outperforms the other two networks, when a sufficiently large number of training examples is available and a relatively large embedding length is chosen. We stress that these findings are obtained under controlled, synthetic conditions with a small number of columns and it should therefore be regarded as an initial, narrowly-scoped investigation rather than a general characterization of ITP on real-world tabular data.
Low-dose computed tomography (LDCT) measurements contain mixed Poisson-Gaussian noise. However, most self-supervised methods rely on generic image statistics and do not explicitly model this noise, which may limit their ability to effectively suppress realistic LDCT noise. To address this issue, we propose a physics-driven framework with cross-domain iteration for self-supervised LDCT denoising. The proposed framework proceeds in three main steps. First, a learned sinogram prior and the LDCT noise model guide posterior inference of photon counts, enabling separation of the Poisson and Gaussian components. Second, the separated Poisson and Gaussian components are respectively processed by binomial thinning and Gaussian data thinning to construct two branches, and residual scaling matches each branch's noise level to that of the observation, yielding a training pair with approximately independent noise realizations from one low-dose measurement. Finally, the pair is used to train an image-domain network whose forward-projected outputs update the prior. Through cross-domain iteration, the prior and the training pair are progressively refined while maintaining consistency with CT acquisition physics. Experiments on simulated data from AAPM, LIDC-IDRI, and LoDoPaB-CT and on real LDCT data show consistent gains over the evaluated self-supervised baselines across dose levels, with performance comparable to the evaluated supervised baseline.
Nikos Giakoumoglou, Andreas Floros, Kleanthis-Marios Papadopoulos +1cs.CV cs.AI cs.LG
We introduce ViTAMINS, a method that integrates synthetic hard negatives into unsupervised vision transformer pretraining to improve representation quality. Our approach is thoroughly benchmarked on ImageNet and transfer learning, image retrieval, copy detection, and image, video segmentation tasks. Notably, our proposed negatives give rise to emergent properties, where learned representations contain explicit information about the semantic content of an image and serve as excellent classifiers (up to +11.3% over baselines). ViTAMINS achieves these benefits through simple modifications to existing contrastive frameworks and outperforms competing methods while being more resource efficient, e.g., our ViT-B surpasses V-JEPA with ViT-L. Our findings motivate reconsidering contrastive learning as a simpler yet powerful alternative to dominant generative and self-distillation approaches.
Samuel Young, César Jesús-Valls, Kazuhiro Teraohep-ex cs.CV
Foundation models are increasingly being pursued in particle and nuclear physics, but existing approaches remain strongly tied to individual experiments through detector-specific architectures or pre-training objectives, limiting their reuse across sensing modalities. We show that a point cloud self-distillation framework yields a substantially more general sensor-level pre-training recipe. We show that the same refined architecture and objective can be independently pre-trained with minimal changes on three qualitatively different detector modalities: liquid argon time projection chamber (LArTPC), collider TPC, and water Cherenkov. Using 1,000 labeled images for downstream task adaptation, Panda V2 matches or exceeds specialized foundation-model baselines trained with orders of magnitude more supervision, matching state-of-the-art particle-clustering performance with 70x fewer labeled events on sPHENIX while substantially improving particle identification, and on LArTPC data matching Panda (arXiv:2512.01324) particle reconstruction with up to 1,000x fewer labels. Beyond reconstruction, simple linear probes reveal physically meaningful latent structure associated with particle causality and track curvature.
Video-text models adapted from image-text architectures (e.g., CLIP) frequently exhibit temporal blindness, the inability to perceive fundamental cues like order, direction, and motion dynamics. Standard datasets mask this limitation by enabling models to exploit static spatial shortcuts. To systematically evaluate this, we introduce XTE-Bench, a diagnostic probe revealing that even large-scale video-language models struggle with basic temporal reasoning, indicating that parameter scaling alone is insufficient to resolve this flaw. To address this, we propose Cross-Modal Temporal Edits (XTE), a self-supervised framework that injects precise temporal supervision. By performing synchronized video-text transformations, XTE generates hard temporal negatives without manual annotation. We instantiate this with ViTAL-X, a lightweight model that equips frozen image-text backbones with temporal awareness while preserving their foundational spatial knowledge. Across six temporal benchmarks, ViTAL-X achieves state-of-the-art performance. Utilizing only 0.4B parameters and 1M training clips, ViTAL-X outperforms 7B-parameter models and surpasses baselines trained on 600x more data. These results demonstrate that targeted, high-quality temporal alignment provides a highly efficient alternative to pure scaling.
Speech emotion recognition (SER) faces two fundamental challenges: scarcity of labeled data and inter-speaker variability, both of which hinder generalization of emotion recognition systems. While prior adversarial approaches address speaker variability, they fall short in leveraging powerful pre-trained representations. We propose SISER (Speaker-Invariant Speech Emotion Recognition), integrating wav2vec 2.0 as a feature encoder and ECAPA-TDNN as a speaker discriminator within an entropy-based adversarial training scheme. wav2vec 2.0 provides rich self-supervised representations that alleviate dependency on large labeled datasets, while ECAPA-TDNN enables suppression of speaker identity via a stronger adversarial signal than shallow classifiers. Evaluated on IEMOCAP, SISER achieves a UA of 60.63%, outperforming the baseline (51.15%) and wav2vec 2.0 without speaker suppression (56.46%), with ablation emphasizing that the choice of speaker classifier architecture is a key factor.
Athira J. Jacob, Puneet Sharma, Dorin Comaniciu +1cs.CV cs.AI
Cardiac magnetic resonance imaging (CMR) produces rich sequential data such as temporal cine videos and spatial LGE/mapping stacks, yet most deep learning approaches process individual 2D slices, discarding this context. We present MR-JEPA, a self-supervised video foundation model for CMR that extends LeJEPA to 3D spatiotemporal inputs through tubelet tokenization, spatiotemporal masking augmentation, and initialization from a 2D CMR foundation model. Unlike prior CMR video models limited to cine data, MR-JEPA is pretrained on multi-sequence data (cine, LGE, mapping) from 10,505 patients across two centers without annotations. We evaluate the frozen encoder on six downstream tasks using a unified multi-view gated attention architecture: LV ejection fraction, RV ejection fraction, three myocardial strains (GLS, GCS, GRS), and four-class disease detection. MR-JEPA outperforms other compared methods on all five regression tasks, including both a domain-specific CMR model pretrained on more data with text supervision and a natural-video foundation model, achieving an LV EF MAE of 4.79% (r =0.764) and a GLS MAE of 1.87 (r=0.805), with 21-27% MAE reductions over baselines on strain tasks. For disease detection, MR-JEPA achieved a macro AUG of 0.868, remaining competitive with the domain-specific baseline despite using a fully self-supervised pretraining objective. These results demonstrate the potential of a unified video encoder for robust, multi-view utilization of diverse CMR sequences in clinical cardiac quantification and diagnosis.
Gabriel Meseguer-Brocal, Yuexuan Kong, Romain Hennequincs.SD cs.AI cs.LG eess.AS eess.SP
Joint-Embedding Predictive Architecture (JEPA) has shown strong performance in learning rich representations through self-supervised prediction in latent space. However, it typically relies on teacher--student architecture with an EMA to stabilise training, and can tend to yield uninformative representations. Contrastive learning is stable to train and produces strong global representations, but remains limited on local tasks by the global nature of its objective. In this work, we combine both into CoJEPA: a single shared backbone jointly trained with a JEPA objective on masked sequence tokens and a contrastive objective on the class token. The contrastive gradient provides stability, removing the need for an EMA teacher entirely, while JEPA enriches the sequence tokens via local predictions that contrastive learning alone cannot provide. Crucially, no extra parameters are added to the backbone: the same model is guided towards richer representations purely through the design of its training signal. CoJEPA takes the best of both worlds, outperforming or matching both individual methods across global and local MIR tasks, with a particularly strong advantage on tonal and harmonic understanding, and without any task-specific architectural changes. CoJEPA shows that combining objectives with complementary inductive biases can substitute for scale, encouraging future work to invest in smarter training objectives over ever-larger models.
We compare six self-supervised pretext tasks for infant cry analysis under a fixed budget, meaning the same compact encoder of 1.17M parameters, the same 115 hours of license-verified public pretraining audio, and the same evaluation protocol for every candidate. On cry detection the reconstructive objectives dominate, and a linear probe over a masked-spectrogram encoder reaches 0.988 AUC with subject-wise splits even though the encoder never observed a cry during pretraining. On cry-reason classification over donateacry, the de facto public benchmark for cry reasons, every encoder performs at chance (0.38 to 0.54 macro AUC over 5 classes), and neither domain adaptation on 1.8 hours of real cries nor end-to-end fine-tuning moves the result. Since a frozen HuBERT-base with 80 times more parameters shows the same pattern, the bottleneck must sit in the labels and not in model capacity. We then reproduce the 90\%+ accuracies of the donateacry literature on our own system by changing nothing but the evaluation protocol: clip-wise splits raise accuracy to 85.2% (barely above the 83.8% majority-class baseline), and applying augmentation before splitting raises it to 97.9%, matching the reported state of the art, from the same model that measures 0.49 macro AUC under subject-wise splits. Under leakage-free splits, a twentyfold augmentation of the labeled set (vocoder speaker perturbation and noise mixing, 21 hours) leaves cross-subject AUC unchanged: for this task the effective sample size is the number of infants. We release code, seeds and per-clip license manifests.
Point cloud video representation learning is crucial for 3D dynamic scene understanding. In this paper, we propose MoSaiC, a novel Motion-Saliency Complementary masked modeling framework for self-supervised point cloud video representation learning. MoSaiC couples three components: Curriculum Motion-Saliency Masking (CMSM), which guides the masking process toward motion-salient tokens under a curriculum schedule; Normal-Flow Motion (NFM) modeling, which supervises the local rigid rotation of each token in the Lie algebra so(3) as an explicit geometric motion target; and Cross-view Token Consistency Prediction (CTCP), which enforces consistency between two complementary masked views at the token level. Together, these components allow MoSaiC to effectively capture both appearance and motion dynamics. Extensive experiments on multiple downstream tasks, including action recognition, temporal action segmentation, and point-level semantic segmentation, demonstrate the effectiveness of our approach.
David Sulu, Lorenzo Di Fruscia, Jana M. Webercs.LG q-bio.BM
Molecular self-supervised learning uses chemical structures to guide which molecular embeddings should be similar. We study whether explicitly encoding a molecule's Bemis-Murcko scaffold and using it to supervise the molecular embedding changes what the model learns. We further test whether this effect depends on the embedding geometry by comparing Euclidean and Lorentz contrastive objectives. Across two augmentation strengths, scaffold-supervised models consistently organize molecules according to both identical and structurally related scaffolds. The resulting embeddings also improve molecular property prediction on several tasks, while the exact gains depend on the predicted property. The effect of scaffold supervision on molecular organization is stronger under Lorentz objectives, but neither geometry provides a consistent overall advantage. These results show that explicitly teaching the relation between a molecule and its structural core can reliably shape the organization of molecular embedding space, while the extent of usefulness of this organization remains task dependent.
Much of human health and function unfolds beyond the clinic, through the movements of everyday life. Wrist-worn accelerometers capture these movements continuously, yet their rich signals are often reduced to a small set of predefined behavioural summary measures. Here, we present Sensori, a self-supervised foundation model that learns general-purpose health representations directly from 24 hours of raw tri-axial wrist movement. We developed and evaluated the model across four population-based cohorts from the United Kingdom, China and the United States, comprising 122,640 participants contributing 683,617 person-days of free-living recordings. Sensori condensed each day of movement into a representation that captured diverse movement behaviours, demographic characteristics, health axes and physical function. Evaluation in independent cohorts showed that these representations generalised across populations and measurement settings without retraining. When added to common clinical covariates, Sensori significantly improved prevalent disease classification for 52 of 102 eligible conditions (median delta AUROC, 0.060; range, 0.012-0.242) and incident disease risk prediction for 26 of 87 eligible conditions (median delta Uno's C-index, 0.064; range, 0.025-0.172), with the largest gains for neurological and psychiatric disorders. These findings establish 24-hour wrist movement as a rich and scalable source of health information, with the potential to support passive health monitoring and disease prediction at population scale.
Alexander Rusnak, Sophia Kovalenko, Jingru Wang +3cs.CV cs.AI cs.LG
Reliable semantic representations derived from city-scale 3D models are increasingly important for urban analysis, infrastructure monitoring, autonomous systems, and heritage conservation. However, urban scenes of large spatial extent captured through aerial surveying differ substantially from the indoor, object-level, and self-driving LiDAR data used to pretrain most 3D self-supervised models. We introduce Polis, to our knowledge the first application of Sketched Isotropic Gaussian Regularization (SIGReg) as an objective for a native point cloud encoder, and evaluate it through a frozen-feature benchmark spanning fourteen city- and building-scale corpora. Polis combines geometrically matched cosine invariance, SIGReg, and VICReg-style anti-collapse terms with a 12.8k-scene outdoor pretraining mixture and gravity-preserving spatial view sampling. Controlled ablations show that this objective outperforms student--teacher architecture alternatives, as well as Polis versions without anti-collapse terms, on the same representative outdoor corpus. On three pretraining-disjoint city datasets, Polis reaches $23.8\%$ mean mIoU versus $16.3\%$ for the next-best encoder under high-capacity frozen probing, and $17.3\%$ versus $16.1\%$ at a matched point and voxel budget. The same city-scale lead holds on datasets whose training sets were seen in pretraining. On localized terrestrial captures with fine-grained facade and streetscape labels, the ranking reverses. Our results show that distributionally-regularized joint embedding architectures can be successful on challenging city-scale 3D scenes, and that transfer improves when self-supervision is designed for the capture geometry and spatial context of this domain while also revealing the limits of this specialization.
Louis Brusset, Mathurin Petit, Jordan Kam +1cs.LG eess.AS
As air traffic volumes in the National Airspace System continue to expand, in particular in the low altitude airspaces, the need for scalable decision support tools used by air traffic controllers will also require more development. This article introduces Voice-to-Trajectory for Air Traffic Control, a joint voice communication-flight trajectory data embedding framework, that can be a component of situational awareness in congested airspaces, and assist the development of tools for ATC as they reason in real-time over Automatic Dependent Surveillance-Broadcast trajectories, or the intent expressed by pilots in natural language. We show that these data modalities are not independent and represent a common physical referent: an aircraft flying through the airspace. V2TATC maps a voice instruction and the trajectory of the addressed aircraft to nearby points in a single latent space that can be queried in both directions. It combines a self-supervised trajectory encoder, a frozen large-scale speech encoder, a contrastive joint embedding, and a bijective lifting via normalizing flows. We demonstrate V2TATC's effectiveness on the San Francisco Bay Area, for its concentration of major airports, and its mix of commercial and general aviation low altitude traffic. Lastly, we release a novel paired voice-trajectory dataset, and report experiments on cross-modal retrieval, ablations, and latent-space analysis.
Galo Castillo-López, Alexis Lombard, Gaël de Chalendar +1cs.CL cs.LG
We propose a multi-task learning approach for multi-party dialogue intent recognition that leverages an auxiliary task that models turn-taking dynamics. Specifically, we introduce turn-transition entropy, a self-supervised target computed from the sequence of speaker transitions, which quantifies the predictability of interaction patterns. Experiments on multiple pre-trained models demonstrate that incorporating this auxiliary task improves intent recognition performance, outperforming existing approaches which ignore multi-party interaction dynamics. We find that our proposed continuous target can be learned as a single-task objective, suggesting that it is an actual signal carrying useful information.
Joint-embedding predictive architectures (JEPAs) have primarily been developed for self-supervised representation learning. Denoising JEPA (D-JEPA) recently demonstrated strong generative capabilities on natural images, yet the applicability to 3D medical imaging remains unexplored. Building on the D-JEPA framework, we present Med-D-JEPA, a systematic adaptation and evaluation of joint-embedding predictive generation for 3D brain MRI. Med-D-JEPA operates on continuous latent tokens produced by a 3D KL-regularized adversarial variational autoencoder, and combines masked context prediction, representation-level alignment, per-token diffusion, and iterative next-set-of-token sampling. We evaluate unconditional and class-conditional generation quality on BraTS2019 and OASIS-1 datasets; downstream classification utility; and preliminary whole-tumor segmentation on BraTS2020. Across different generation settings, Med-D-JEPA achieves superior or competitive performance compared to several strong baselines on fidelity and diversity metrics. Compared to training with real samples, Med-D-JEPA-based synthetic pretraining improves classification AUC from 0.63 to 0.85 on BraTS2019 and from 0.78 to 0.87 on OASIS-1. In the segmentation study, pretraining on Med-D-JEPA samples improves Dice from 0.74 to 0.80 and reduces HD95 from 13.40 to 9.56 mm. These findings establish joint-embedding predictive generation as a promising direction for 3D medical image synthesis and encourage further research in this direction.
V. S. D. S. Mahesh Akavarapu, Michael Daniel, Gerhard Jägercs.CL
Forced alignment evaluation typically requires manually annotated timestamps, limiting large-scale and multilingual analysis. We introduce two corpus-level metrics based on self-supervised (SSL) speech representations for reference-free forced alignment evaluation: Phoneme-Cluster Mutual Information (PCMI) and Word Acoustic Consistency Score (WACS). PCMI measures agreement between aligned phoneme labels and clusters induced from SSL-speech representations, while WACS measures consistency of repeated word realizations using dynamic time warping similarity between word representation sequences. Using both random and systematic perturbations, we show that PCMI and WACS degrade consistently under alignment perturbations. We further analyze the metrics across multiple alignment systems on 85 languages from FLEURS, validate them against manually annotated alignments from 45 languages in DoReCo, and evaluate them on two phonologically complex low-resource languages. The metrics effectively separate high- and low-quality alignments and correlate strongly with timestamp-based alignment quality measures. Our results demonstrate that SSL-speech representations enable scalable, reference-free forced alignment evaluation. The metrics are available as an open-source Python package at https://github.com/mahesh-ak/forced-aligner-metrics.
Lukas Kuhn, Lucas Maes, Giuseppe Serra +4cs.CV cs.AI
Video carries the temporal structure of the physical world, yet learning representations from it has remained computationally expensive: prevailing self-supervised methods either prevent representation collapse through architectural asymmetries, coupling an exponential-moving-average target encoder, a stop-gradient, and a capacity-limited predictor, or circumvent it by reconstructing masked content in pixel space. We introduce LeVJEPA, the first video encoder trained under LeJEPA's collapse-free objective, which dispenses with both. A single encoder is trained with an invariance loss over global and local views of a clip, regularized by SIGReg, which excludes collapse with a provable guarantee. The architecture reduces to an encoder and a projector, and the objective to a single hyperparameter. This formulation admits two properties. First, the cost of pretraining is governed by the number of tokens the encoder observes; uniform random token dropping renders this number small while simultaneously improving downstream accuracy. At matched epochs on identical data, LeVJEPA matches or surpasses V-JEPA 2 across ViT-S/B/L at 5.6 to 20.8x less pretraining compute, and at matched total FLOPs it exceeds the strongest video baseline by 7.6 points on ImageNet-1K while remaining competitive on motion-centric benchmarks. Second, since no asymmetry between branches is required, the encoder can be trained with block-causal attention at no measurable accuracy cost: temporal ordering becomes a property of the encoder itself. Against a compute-matched DINOv2 trained on frames of the same videos, LeVJEPA approaches the image-pretrained encoder on appearance-centric evaluation while nearly doubling its motion-centric accuracy. These results indicate that, once its computational overhead is removed, video becomes a viable and in several respects preferable substrate for general-purpose visual pretraining.