Self-Supervised

Existing methods for expressive music performance rendering, a conditional generation task that aims to generate a human-like performance from a symbolic score, rely on supervised learning over small labeled datasets, which limits scaling of both data volume and model size, despite the availability of vast unlabeled music, as in vision and language. To address this gap, we introduce Pianist Transformer, with three key contributions: 1) introducing large-scale self-supervised learning into expressive piano performance rendering through a unified Musical Instrument Digital Interface (MIDI) representation, enabling pre-training on 10B tokens of unlabeled MIDI data; 2) an efficient asymmetric Transformer with note-level compression, substantially improving training efficiency, memory usage, and inference speed for long-context music modeling; 3) a state-of-the-art rendering model with an editable workflow, achieving strong objective and subjective results and enabling integration into real-world music production workflows. Overall, Pianist Transformer outlines a scalable path toward human-like performance synthesis in the music domain. Code, audio samples, and model checkpoints are available on our project page: https://yhj137.github.io/pianist-transformer-demo/.

Monocular 3D face reconstruction estimates a 3D morphable model (3DMM) representation from a single image, providing geometry-aware expression codes that are useful for facial expression analysis and affect understanding. Despite strong progress, most pipelines are trained with image-level self-supervision and evaluated primarily by geometric fidelity, which does not necessarily maximize the affective utility of the learned expression representation and may encourage intensity-amplifying shortcuts when affect supervision is naively coupled. We propose FIELDS (Face reconstruction with accurate Inference of Expression using Learning with Direct Supervision), a task-driven framework that learns FLAME expression codes for facial expression recognition (FER) under a geometric plausibility constraint. Using hybrid 2D/3D supervision, FIELDS improves affect prediction in both in-domain and external evaluations while maintaining competitive geometric fidelity on held-out and out-of-domain 3D benchmarks.

Training deep networks with noisy labels leads to poor generalization and degraded accuracy due to overfitting to label noise. Existing approaches for learning with noisy labels often rely on the availability of a clean subset of data. By pre-training a feature extractor on the target dataset without labels using in-domain self-supervised learning (SSL), followed by standard supervised training on the same noisy dataset, we can train a more noise robust model without requiring a subset with clean labels. We evaluate both contrastive and non-contrastive SSL pre-training methods across datasets with synthetic and real-world label noise, demonstrating the broad applicability of our approach across large-scale datasets, diverse downstream tasks, and model architectures. Across all noise rates, in-domain self-supervised pre-training consistently improves classification accuracy and downstream label-error detection (F1 and Balanced Accuracy) compared with supervised training from scratch. The performance gap widens as the noise rate increases, demonstrating improved robustness. Notably, our approach achieves comparable results to ImageNet and DinoV2 pre-trained models at low noise levels, while substantially outperforming them under high noise conditions.

The pretrain-transfer paradigm, which underpins the success of large language models (LLMs), has demonstrated the immense power of creating foundation models that learn generalizable representations from vast datasets. However, extending this paradigm to Operations Research (OR) problems on graph structures remains challenging due to the fundamental conflict between the statistical flexibility of language and the strict combinatorial constraints of graphs. To bridge this gap, we introduce the Graph Foundation Model (GFM), the first framework capable of solving all distance-based optimization problems on graph structures. By introducing the LLM-like self-supervised pre-training paradigm on the paths generated from random walks in the graph, GFM is compelled to internalize the graph's complex topological and combinatorial rules, where the connectivity of the structure itself can be treated as the supervisory signal. Unlike existing neural methods that learn complex and task-specific solving policies, our approach leverages the pre-trained GFM as a foundational model of the graph's intrinsic structure, which in turn enables a simple generative heuristic to tackle a diverse range of optimization challenges effectively. Comprehensive experiments on networks ranging from 20 to 893 nodes demonstrate that GFM achieves competitive performance against specialized solvers across a variety of distinct optimization task classes, while maintaining significantly faster inference times. Our work establishes a new paradigm of adapting the pretrain-transfer framework to graph optimization, opening the door for applying foundation model innovations to OR.

Current deep learning models for electroencephalography (EEG) are often task-specific and depend on large labeled datasets, limiting their adaptability. Although EEG foundation models seek broader applicability, many still rely on predefined multi-channel inputs, electrode-layout assumptions, or model-specific channel handling. To address these limitations, we introduce the Single-Channel Large EEG Model (SingLEM), a self-supervised foundation model whose hybrid convolutional--Transformer encoder maps each channel independently to a reusable representation capturing local and long-range temporal structure. These representations can be used individually or combined through channel-wise feature concatenation. We assembled 71 public EEG datasets comprising approximately 9,200 subjects and 357,000 single-channel hours. For leakage-controlled evaluation, downstream results were obtained with a model pretrained on 68 datasets after excluding the three source datasets underlying the six tasks. A model pretrained on all 71 datasets is provided for general reuse. Across six motor imagery and cognitive tasks under strict leave-one-subject-out (LOSO) evaluation, the leakage-controlled model with concatenated representations and a support vector machine (SVM) classifier achieved the best overall performance among the compared pretrained and classical feature-based methods. Additional classifier and subject-adapted analyses supported the robustness of its representations. These findings support single-channel self-supervised learning as a montage-flexible foundation for reusable EEG feature extraction and electrode-level spatial analysis. The source code and pretrained models are available at https://github.com/ttlabtuat/SingLEM.

For groups of autonomous agents to achieve a particular goal, they must engage in coordination and long-horizon reasoning. Rather than relying on complex reward functions and explicit cooperation mechanisms, we ask what minimal ingredients are required for effective coordination and exploration to emerge in multi-agent settings. We investigate this question through self-supervised goal-reaching, where agents aim to maximize the likelihood of visiting a goal state rather than maximizing a reward. Despite a sparse feedback signal, we present empirical results that show self-supervised goal-reaching techniques enable agents to learn from such feedback. On MARL benchmarks, self-supervised goal-reaching outperforms alternative approaches that have access to the same sparse reward signal. Furthermore, we empirically demonstrate that multi-agent self-supervised goal-reaching approaches can be more robust than single-agent strategies. While there is no explicit exploration mechanism, this approach explores nontrivial intermediate coordination strategies in sparse settings where alternative approaches fail to achieve a single success.

Self-supervised learning has become an increasingly important paradigm in the domain of machine intelligence. Furthermore, evidence for self-supervised adaptation, such as contrastive formulations, has emerged in recent computational neuroscience and brain-inspired research. Nevertheless, current work on self-supervised learning relies on biologically implausible credit assignment -- in the form of backpropagation of errors -- and feedforward inference, typically a forward-locked pass. Predictive coding, in its mechanistic form, offers a biologically plausible means to sidestep these backprop-specific limitations. However, unsupervised predictive coding rests on learning a generative model of raw input (akin to "generative AI" approaches), which entails predicting a potentially high dimensional input; on the other hand, supervised predictive coding, which learns a mapping between inputs to target labels, requires human annotation, and thus incurs the drawbacks of supervised learning. In this work, we present a scheme for self-supervised learning, specifically for an emerging research sub-domain that we label as neuroscience-informed self-supervised learning (NeuroSSL), within a neurobiologically plausible framework that appeals to the free energy principle, constructing a new form of predictive coding that we call meta-representational predictive coding (MPC). MPC sidesteps the need for learning a generative model of sensory input (e.g., pixel-level features) by learning to predict representations of the input across parallel streams, resulting in an encoder-only learning and inference scheme. This formulation notably rests on active inference (in the form of sensory glimpsing) to drive the learning of representations, i.e., the representational dynamics are driven by sequences of decisions made by the model to sample informative portions of its sensorium.

Hyperspectral imaging is a powerful bioimaging tool which can uncover novel insights, thanks to its sensitivity to the intrinsic properties of materials. However, this enhanced contrast comes at the cost of system complexity, constrained by an inherent trade-off between spatial, spectral, and temporal resolution. To overcome this limitation, we present a self-supervised deep learning-based approach that restores and enhances pixel resolution post-acquisition without requiring external training data beyond the images to be restored. Fine-tuned using metrics aligned with the imaging model, our physics-aware method achieves a 16$\times$ pixel super-resolution enhancement and a 12$\times$ imaging speedup without the need of additional training data for transfer learning. Applied to both synthetic and experimental data from five different sample types, including healthy and diseased tissues, we demonstrate that the model preserves biological integrity, as we did not detect systematic loss of biological features or biologically consequential hallucinations in tested datasets. We also concretely demonstrate the model's ability to reveal disease-associated metabolic changes that would otherwise remain undetectable. Furthermore, we provide physical insights into the model's inner workings, paving the way for future refinements that could potentially reveal novel high resolution features in an explainable manner. All methods are available as open-source software on GitHub.

We introduce the novel concept of visually Connecting Actions and Their Effects (CATE) in video understanding. CATE can have applications in areas like task planning and learning from demonstration. We identify and explore two different aspects of the concept of CATE: Action Selection (AS) and Effect-Affinity Assessment (EAA), where video understanding models connect actions and effects at semantic and fine-grained levels, respectively. We design various baseline models for AS and EAA. Despite the intuitive nature of the task, we observe that models struggle, and humans outperform them by a large margin. Our experiments show that in solving AS and EAA, models learn intuitive properties like object tracking and pose encoding without explicit supervision. We demonstrate that CATE can be an effective self-supervised task for learning video representations from unlabeled videos. The study aims to showcase the fundamental nature and versatility of CATE, with the hope of inspiring advanced formulations and models.

Autonomous systems (AS) often rely on Deep Neural Network (DNN) classifiers to operate in complex and dynamically changing environments. However, during operation, these classifiers may encounter domains that differ from those seen during development, causing performance degradation under distribution shift. Removing systems from operation for labelled data collection and retraining is often impractical, particularly when adaptation must occur quickly and at scale. This paper introduces DIRA-SS, a self-supervised extension of Dynamic Incremental Regularised Adaptation (DIRA) that enables online domain adaptation using only a small number of unlabelled target-domain samples. DIRA-SS augments an existing classifier with an auxiliary retraining branch and adapts the shared feature representation through a rotation-prediction task, while elastic weight consolidation regularises important source-domain parameters to reduce destructive updates. This allows the model to benefit from transfer learning without requiring classification labels during operation. We evaluate DIRA-SS on CIFAR-10C, CIFAR-100C, and ImageNet-C using ResNet architectures under severe common corruptions. The results show that DIRA-SS substantially improves performance over the non-adapted source model, achieves accuracy close to the supervised DIRA method, and outperforms existing unsupervised test-time adaptation baselines on ImageNet-C when using only 100 target-domain samples.

Ptychographic imaging at synchrotron and X-ray free-electron laser sources requires densely overlapping scans, which limits throughput and increases dose; extending coherent diffractive imaging to overlap-free operation on extended samples remains an open problem. We present a self-supervised inverse-mapping network for single-frame Fresnel coherent diffraction imaging (CDI) and overlapped ptychography with fixed, pre-estimated probes. The learned neural network reconstructs individual object patches from either one diffraction frame or several overlapping measurements at a time. In single-frame mode, the phase diversity provided by the curved-wavefront probe at the off-focus sample position removes the requirement for overlap constraints, enabling sparser scans and proportionally lower dose at fixed exposure. On synthetic line patterns, reconstructed amplitude SSIM exceeds 0.90 in single-frame mode with the curved probe and reaches 0.952-0.968 with overlap constraints. Optimization of the network via a Poisson negative log likelihood objective, rather than the more common mean absolute error, yields 10-fold improved photon-dose efficiency at doses below $10^5$ photons per image, where shot noise typically limits resolution. In addition to these synthetic studies, we demonstrate robust single-frame reconstruction of extended samples using ptychographic datasets from APS and LCLS, with end-to-end reconstruction of a 10,304-frame workload approximately $36\times$ faster than a highly optimized iterative solver. Together, these results unify single-frame Fresnel CDI and overlapped ptychography within one self-supervised framework, supporting dose-efficient, high-throughput imaging at modern light sources.