Transfer Learning

This paper proposes a novel approach to address the challenge that pretrained VLA models often fail to effectively improve performance and reduce adaptation costs during standard supervised finetuning (SFT). Some advanced finetuning methods with auxiliary training objectives can improve performance and reduce the number of convergence steps. However, they typically incur significant computational overhead due to the additional losses from auxiliary objectives. To simultaneously achieve the enhanced capabilities of auxiliary training with the simplicity of standard SFT, we decouple the two objectives of auxiliary-objective SFT within the parameter space, namely, enhancing general capabilities and fitting task-specific action distributions. To deliver the goal, we only need to train the model to converge on a small-scale task set using two distinct training strategies, resulting in two finetuned models. The parameters' difference between the two models can then be interpreted as capability vectors provided by auxiliary objectives. These vectors are then merged with pretrained parameters to form a capability-enhanced meta model. Moreover, when standard SFT is augmented with a lightweight orthogonal regularization loss, the merged model attains performance comparable to auxiliary finetuned baselines with reduced computational overhead. Internal and external experiments demonstrate that our capability vectors (1) are effective and versatile across diverse models, (2) can generalize to novel environments and embodiments out of the box.

Intracranial electrocorticography (ECoG) offers high signal-to-noise access to cortical activity for brain-computer interfaces, yet limited per-patient data has led most prior work to rely on small, subject-specific decoders that neglect information shared across patients. We investigate whether large pretrained scalp-EEG foundation models (EEG FMs) can be adapted to ECoG, enabling cross-patient learning and competitive decoding performance while calibrating to a held-out patient in 10-30 minutes on a single GPU. We introduce CORTEG, a cross-modality transfer framework that combines a pretrained EEG FM backbone, an electrode-aware KNNSoftFourier spatial adapter, a dual-stream tokenizer for low-frequency and high-gamma activity, and a leave-one-subject-out fine-tuning strategy. We evaluate CORTEG on two challenging regression tasks: public finger-trajectory regression ($n=9$) and private audio-envelope regression ($n=16$). CORTEG achieves state-of-the-art mean performance on both tasks, achieving the highest mean correlation among the compared methods on the public finger benchmark and larger, statistically significant gains on the audio task and in low-data-per-patient calibration. We further validate CORTEG on the public BrainTreebank stereo-EEG classification tasks, where it also performs strongly. Furthermore, CORTEG outperforms the compared intracranial-EEG foundation models across the continuous regression tasks. Feature analyses align with neurophysiology, and latent representations capture low-dimensional finger-movement structure. Together, these results provide systematic evidence that scalp-EEG pretraining can be repurposed across intracranial recording modalities and tasks, enabling data-efficient neural decoders that can adapt to new patients.

Large language models from different families use different hidden dimensions, tokenizers, and training procedures, making behavioral directions difficult to compare or transfer across models. We introduce an anchor-projection framework that maps hidden representations from each model into a shared anchor coordinate space (ACS). Behavioral directions extracted from source models are projected into ACS and averaged into a canonical direction. For a new model, the canonical direction is reconstructed into its native hidden space using only anchor activations, without fine-tuning or target-specific direction extraction. We evaluate five instruction-tuned model families and ten behavioral axes. We find that same-axis directions align tightly across the Llama-Qwen-Mistral-Phi (LQMP) cluster in ACS. This shared structure transfers to downstream tasks. For the aligned LQMP cluster, held-out targets achieve (0.83) ten-way detection accuracy and (0.95) mean binary AUROC, while canonical steering induces refusal-rate shifts of up to +0.46% under distribution shift. Sensitivity analyses show that two source models and small anchor pools already suffice to approximate transferable directions. Overall, ACS provides a novel perspective on cross-family interpretability, revealing that representation-level transfer remains robust across model families.

We introduce SeePhys Pro, a fine-grained modality transfer benchmark that studies whether models preserve the same reasoning capability when critical information is progressively transferred from text to image. Unlike standard vision-essential benchmarks that evaluate a single input form, SeePhys Pro features four semantically aligned variants for each problem with progressively increasing visual elements. Our evaluation shows that current frontier models are far from representation-invariant reasoners: performance degrades on average as information moves from language to diagrams, with visual variable grounding as the most critical bottleneck. Motivated by this inference-time fragility, we further develop large training corpora for multimodal RLVR and use blind training as a diagnostic control, finding that RL with all training images masked can still improve performance on unmasked validation sets. To analyze this effect, text-deletion, image-mask-rate, and format-saturation controls suggest that such gains can arise from residual textual and distributional cues rather than valid visual evidence. Our results highlight the need to evaluate multimodal reasoning not only by final-answer accuracy, but also by robustness under modality transfer and by diagnostics that test whether improvements rely on task-critical visual evidence.

Preliminary low-thrust spacecraft mission design is a global search problem characterized by a complex solution landscape, multiple objectives, and numerous local minima. During this phase, mission parameters are often not yet fully defined, requiring new solutions to be generated at a high cadence across varying parameter values. When combined with the indirect approach to optimal control, diffusion models can accelerate this search by learning distributions that represent high-quality initial costates. However, generating training data remains expensive, and opportunities exist to better exploit past data. We propose a transfer-learning framework that combines homotopy in a mission parameter with Markov chain Monte Carlo (MCMC) to generate training data more efficiently. The approach reformulates a multiobjective optimization problem as sampling from an unnormalized target distribution in costate space. We compare three MCMC algorithms on a planar multi-revolution transfer in the circular restricted three-body problem, with homotopy in the system mass parameter. The results show that gradient-based MCMC variants achieve the best trade-off between sample quality and computational cost. For the test transfer, the proposed framework generates 40 % more feasible solutions and achieves a higher-quality Pareto front than a state-of-the-art indirect approach based on adjoint control transformations and gradient-based optimization. Finally, the MCMC-generated samples are used to fine-tune a diffusion model conditioned on the mass parameter, enabling it to learn a global representation of the underlying solution distribution and efficiently generate new solutions. These findings establish the transfer-learning framework as a practical method for efficiently solving indirect trajectory optimization problems with varying parameters.

Transfer learning under limited data is a challenging setting, where models must adapt to new tasks with minimal supervision. Prior work has primarily focused on improving absolute accuracy in transfer learning. However, empirical evidence comparing quantum and classical models in realistic transfer learning settings remains limited, especially in low-data regimes. In this work, we systematically study the robustness of quantum models under reduced training data. We evaluate multiple quantum and classical architectures across diverse transfer tasks and retraining configurations, and quantify robustness using accuracy degradation and relative performance retention (RPR). Our results show that, although classical models often achieve higher peak performance, they exhibit significantly larger degradation when training data is limited. In contrast, quantum models maintain more stable performance across data regimes, indicating improved robustness and data efficiency. These findings provide empirical evidence that quantum models can offer improved robustness in low-resource transfer learning scenarios.

Purpose: To develop and evaluate cross-sequence transfer learning for automatic femoral cartilage segmentation, testing bidirectional transfer between dual-echo steady-state (DESS) and sagittal proton density-weighted 3D fast spin-echo (Cube) sequences. Materials and Methods: We optimized a modified 2D U-Net on 507 DESS images from the Osteoarthritis Initiative (OAI). We then established same-sequence baselines using subject-level cross-validation on a subset of 44 OAI DESS images and 44 Cube images acquired at the Istituto Ortopedico Rizzoli, Bologna, Italy. Each subset included 22 non-lesioned and 22 lesioned subjects. Finally, we performed transfer learning across sequences by fine-tuning the pretrained models on the target sequence with increasing training set sizes to study convergence, while keeping validation and test sets fixed. Segmentations were evaluated using Dice similarity coefficient (DSC) and average surface distance (ASD). Lesion effects were assessed with two-sided Mann-Whitney U tests with Bonferroni correction. Results: Same-sequence training yielded higher accuracy on DESS than Cube (DSC, $0.900$ vs $0.830$; $P < .001$). Cube-to-DESS transfer matched DESS performance (DSC, $0.903 \pm 0.032$ vs $0.900 \pm 0.027$), reaching a performance plateau at 9 training subjects. DESS-to-Cube yielded a lower combined DSC ($0.802 \pm 0.049$ vs $0.830 \pm 0.042$), reaching a plateau at 24 training subjects. Lesions did not affect DESS ($P \ge .39$) but reduced Cube accuracy (DSC, $0.805$ vs $0.856$; $P < .001$). Conclusion: Transfer learning across sequences can substantially reduce target-sequence annotation requirements for femoral cartilage segmentation, but performance is direction- and sequence-dependent, and the effects of lesions on segmentation may vary across MRI sequences.

Crystal generative models mainly learn what stable crystals look like, with little explicit supervision for what makes them stable. We reveal a substantial representation gap between state-of-the-art crystal generative models and pretrained universal machine learning interatomic potentials (MLIPs) via energy probing, and show this gap can be closed by a simple training-time alignment. We propose Crystal REPresentation Alignment (CrystalREPA), a plug-and-play framework that aligns the atom-wise hidden states of generative encoders with frozen MLIP representations through an element-aware contrastive objective, transferring stability-aware atomistic priors with marginal training overhead and no additional inference cost. Across three generative frameworks, ten MLIP teachers, and two benchmark datasets, CrystalREPA consistently improves the thermodynamic stability, structural validity, and structural fidelity of generated crystals. Equally important, we find that an MLIP's transfer effectiveness is poorly predicted by its accuracy on standard leaderboards (e.g., Matbench Discovery) but strongly predicted by the distinguishability of its atom-wise representation space, yielding a practical, accuracy-independent criterion for selecting MLIP teachers for generative transfer. Code is available at https://github.com/Chengqian-Zhang/CrystalREPA.

Deep learning techniques have revolutionised medical imaging, improving diagnostic accuracy and enabling both more accurate and earlier disease detection. However, the relationship between pre-training strategies and downstream performance in medical imaging models requires further exploration. Here, we systematically compare convolutional neural networks and transformers, examining various pre-training approaches, including supervised and self-supervised learning, as well as different initialisations and data modalities. Models are evaluated on natural images, chest X-rays, chest CT and retina OCT images, considering the effects of matching pre-training data with target modalities. Our findings indicate that only pre-training on data closely matching the target modality significantly improves downstream performance. While self-supervised learning can outperform supervised methods, its effectiveness varies with context. The study underscores the importance of pre-training strategies to enhance the reliability and effectiveness of deep learning models in medical imaging. By addressing these key factors, our research aims to contribute to the development of more accurate and dependable diagnostic tools, ultimately improving patient outcomes in clinical settings.

We propose CAST, a dual-stream architecture that utilizes channel-aware spatial transfer learning for isolated sign language recognition addressing the challenges of magnitude-only 60~GHz radar Range-Time Maps (RTM). The proposed framework combines three physics-aware architectures with pretrained vision backbones, which operate under radar-only constraints across clinical and alphabetical gestures. First, an explicit decibel-to-linear inversion is combined with a windowed fast Fourier transform that extracts Cadence Velocity Diagrams (CVD) while avoiding the harmonic artifacts that arise from the spectral analysis of log-compressed signals. Second, a cross-antenna spatial attention module applies attention to raw antenna channels before the convolution, preserving inter-receiver amplitude covariance. Third, an asymmetric cross-attention mechanism fuses representations from parallel ConvNeXt-Tiny (CVD) and EfficientNetV2-S (RTM) backbones. Extensive experiments reveal that the architecture achieves a Top-1 accuracy of 80.5% under 5-fold cross-validation, establishing a 3.3% improvement over the best single-model baseline (77.2%). The findings suggest that physics-aware signal representations form a promising direction for radar-only sign language recognition under constrained sensor modalities. The source code is available at: https://github.com/Shakhoyat/CAST-at-SignEval2026.

This paper proposes a time-warping transfer learning method, a technique for temporally rescaling the learned dynamics of a recurrent neural network (RNN) with a Long Short-Term Memory (LSTM) layer to enable task transfer across fuel moisture classes. Fuel moisture content (FMC) is divided into idealized classes based on characteristic lag time. Large quantities of real-time data are available for 10h fuels from sensors on weather stations, but observations of other fuel classes are sparse in space and time. We use transfer learning to adapt an RNN pretrained on 10h FMC to predict FMC for 1h, 100h, and 1000h fuels. We validate this method using data from a landmark field study conducted in Oklahoma that was used to calibrate the state-of-the-art Nelson fuel moisture model.

Hybrid State-Space models combine Attention with recurrent State-Space Model (SSM) layers, balancing eidetic memory from Attention with compressed fading memory from SSMs. This yields smaller Key-Value caches and faster decoding than Transformers, along with a richer architectural design space. Exploring that design space at scale has so far required training from scratch, a barrier that has kept most large-model Hybrid research within a narrow range of architectures. We introduce Priming, a method that turns Hybrid architecture design from a pre-training problem into a knowledge transfer one. Priming initializes a Hybrid model from a pre-trained Transformer and, through short alignment and post-training phases, recovers downstream quality using less than 0.5% of the source model's pre-training token budget. Priming is agnostic to the source Transformer family (e.g., Qwen, Llama, Mistral), model class (dense or Mixture-of-Experts), and model scale. Priming enables us to run the first controlled comparison of SSM layer types at scale under identical conditions. We evaluate, Gated KalmaNet (GKA), Gated DeltaNet (GDN), and Mamba-2, and show that their expressiveness hierarchy, GKA>GDN>Mamba-2, directly predicts downstream performance on long-context reasoning tasks. We scale Priming to 8B/32B reasoning models with native 128K contexts. Our Hybrid GKA 32B improves over its source Qwen3-32B by +3.8 average reasoning points, while staying within 1% of a Transformer post-trained on the same data and enabling up to 2.3x higher decode throughput. To foster research on Hybrid architectures, we release a model zoo of primed Hybrid models for long-context reasoning and instruction following, together with the Priming training and inference code (Sequence Parallelism algorithms for long-context training, optimized GKA kernels, and vLLM serving plugin), all under Apache~2.0 License.

Reusing the early layers of cohort-trained INRs as initialization for new signals has been shown to accelerate and improve signal fitting, yet it remains unclear which layers of the shared encoder learn transferable representations and what those representations encode. We address both questions for two standard backbones, SIREN and Fourier-feature MLPs (FFMLP). First, sweeping the freeze depth across the shared encoder at test time, we find that the optimum coincides with the layer of highest weight stable rank. Moreover, freezing at this depth matches or improves on the standard fine-tuning recipe across all our experiments. Second, identifying which layer transfers does not characterize what that layer encodes. To address this we adopt sparse autoencoders (SAEs), the dominant tool in mechanistic interpretability, and present the first SAE decomposition of INR activations into sparse dictionary atoms. Interestingly, SIREN and FFMLP achieve comparable cohort-fitting quality, but learn qualitatively different dictionaries. Cohort SIREN's atoms are localized, tiling the coordinate plane such that each atom fires in a confined region independent of cohort content. Cohort FFMLP's atoms are image-spanning, tracing the contours of memorized cohort signals. Single-atom ablations confirm causal use of these dictionaries: a single FFMLP atom out of 4096 can drop PSNR by up to 10.6 dB across the image, while SIREN ablations remain confined to where the atom fires. Together, these results give the first mechanistic account of what transfers in cohort-trained INRs and turn their activations into inspectable dictionary atoms. These tools open a path towards characterizing what INRs encode and towards architectures designed for generalization rather than memorization.

Machine-learning models in high-energy physics are often trained on simulated data, where fully simulated samples are computationally expensive while fast simulation provides large statistics at reduced realism. In this work, we systematically study transfer learning between fast-simulated and fully simulated datasets in a realistic LHC environment. We consider three representative tasks, signal-background classification, quark-gluon jet tagging, and missing transverse energy reconstruction, using dense neural networks, graph neural networks, and transformer-based architectures. Models are pretrained on ATLAS-like fast simulation and adapted to CMS-like fast simulation and to fully simulated ATLAS Open Data. Across all tasks, pretrained models consistently outperform independently trained baselines and require significantly less target-domain training data, typically reducing the needed statistics by about a factor of two. These results demonstrate that fast simulation can be used to learn robust, reusable representations and motivate publishing trained models as reusable scientific assets beyond large foundation models.

Health foundation models (FMs) learn useful representations from wearable sensors, but interpreting what they encode and transferring that knowledge across modalities after training remains difficult. We present a post-training framework that decomposes frozen embeddings into interpretable directions, referred to as symbols, and use these symbols to align the embedding spaces without retraining. We evaluate the framework on three FMs for photoplethysmography (PPG) and accelerometer data, independently pretrained on ~20M minutes of unlabeled data from ~172K participants, and analyzed on a held-out cohort of 30K subjects. We find that extracted symbols associate selectively with health conditions and physiological attributes, and these associations are partially shared across modalities and architectures. Cross-modal transfer via symbols retains more than 95% of in-domain performance, is nearly symmetric across domain directions, and saturates with limited paired data, together indicating that alignment recovers a shared low-dimensional subspace rich in physiological information. Overall, these results suggest that health FM embeddings contain an interpretable symbolic organization that is shared across modalities and supports cross-domain transfer without joint training.

Finetuning pretrained models occurs in a low-dimensional subspace of the full parameter space. Prior work has focused on characterizing this optimization subspace, but largely ignored the complementary question: why do certain directions remain unexplored during finetuning? Are these stable directions irrelevant to downstream tasks, or do they already encode task-relevant structure that requires no further adjustment? Answering this question is central to understanding how pretrained knowledge transfers. Through systematic spectral analysis across vision and language models, we show that the leading singular vectors of pretrained weight matrices remain highly stable under finetuning and are shared across unrelated downstream tasks, revealing that pretraining establishes a reusable spectral coordinate system. Models pretrained on larger datasets exhibit greater spectral stability under distribution shift or task change, directly linking pretraining scale to geometric transferability. Motivated by these findings, we propose a parameter-efficient method that freezes pretrained singular vectors and optimizes only leading spectral coefficients, achieving competitive performance on GLUE with 0.2% trainable parameters. Our results reveal that the stable directions encode transferable structure rather than irrelevant noise: successful pretraining discovers spectral bases that downstream tasks inherit and operate within.

Spiking Neural Networks (SNNs) have gained increasing attention due to their potential for low-power computation on neuromorphic hardware. A widely adopted training strategy for SNNs is direct coding, which enable backpropagation on neuron implementations using continuous-valued surrogate activations. However, recent studies have shown that direct-coded SNNs remain substantially less energy-efficient than their event-based counterparts, limiting their practical deployment in energy sensitive scenarios. Still, to promote the reusability of pretrained SNN database on direct code, this motivates an important yet underexplored question: How can a SNN pretrained with direct code be effectively converted into an event-based representation? In this research, we present the first systematic investigation into this transfer problem, analyze the key challenges that arise when transitioning from direct-coded to event-based computation and propose a set of methods to enable energy-efficient transfer while preserving model performance.

Dataset distillation compresses a large training set into a small synthetic set that preserves downstream training utility. While most existing methods target training networks from scratch, modern visual transfer learning often uses frozen pre-trained encoders followed by lightweight linear probing. Existing distillation methods for this setting either unroll iterative linear-probe updates with trajectory-based gradient matching, or rely on closed-form formulations originally designed for from-scratch training with neural-tangent-kernel (NTK) approximations. Neither route exploits the fact that frozen-feature linear probing admits a closed-form solution determined directly by the pre-trained features themselves, with no infinite-width approximation and no inner-loop trajectory. We propose Closed-Form Linear-Probe Dataset Distillation (CLP-DD), a bilevel formulation that computes the linear probe induced by the synthetic set with a sample-space kernel ridge solver. The synthetic images are then updated by evaluating this induced classifier on real features through a temperature-scaled softmax cross-entropy, where the classifier columns act as learned class anchors in feature space. We further show that the choice of outer objective is decisive: pairing the closed-form inner solver with a standard MSE outer loss substantially underperforms trajectory-based methods, while the discriminative outer loss closes most of the gap. On ImageNet-100 with four pre-trained backbones, CLP-DD substantially improves over LGM without DSA and approaches LGM with DSA at a fraction of the computational cost. On ImageNet-1K, CLP-DD matches or surpasses LGM with DSA on three of four backbones while running roughly $14\times$ faster and using less than one-eighth of the GPU memory.

A recent work shows that Attention Transfer, which transfers only the attention patterns from a pre-trained teacher Vision Transformer (ViT) to a randomly initialized standard student ViT, is sufficient to recover the full benefit of the teacher's pre-trained weights. We revisit this finding on a comprehensive benchmark of 20 teachers from 11 well-known ViT families and reveal that Attention Transfer is not universally effective. While 7 families transfer successfully, 4 consistently fail, falling up to 5.1% below the from-scratch no-transfer baseline. Further results demonstrate that this failure is family-consistent across model sizes, and persists under extended training durations, different transfer datasets, and out-of-distribution evaluations. Controlled analyses then consistently localize the problem to the attention-routing channel, indicating that the key issue is not whether the student can match the teacher's attention patterns, but whether the matched patterns remain functional for the student. Crucially, we identify architectural mismatch between the pre-trained teacher and the standard student as the primary mechanism. By adding only the teacher's native architectural components to the student in a randomly initialized state, we completely reverse the failure for all 4 families. Notably, these components alone do not improve from-scratch training, confirming that they specifically unlock the usability of the teacher's attention. We further show that this failure is not explained by the inadequate choice of transfer loss, differences in pre-training recipes, or the intrinsic quality of attention maps. Our findings refine the prevailing understanding of attention in ViT representations: attention is sufficient only when the student architecture matches the teacher.

Although self-supervised pretraining is expected to learn broadly transferable representations, its effectiveness across imaging modalities substantially different from the pretraining domain, and on complex tumor-segmentation tasks, remains understudied. Evaluating CT-pretrained transformers on MRI rectal cancer segmentation, we identified two interacting failure modes in CT-to-MRI transfer: (a) inefficient token usage caused by zero-padding to match pretrained input dimensions, and (b) ineffective feature adaptation. We investigated these vulnerabilities using two primary CT-pretrained hierarchical shifted-window transformer backbones, SMIT and Swin UNETR, together with VoCo as a large-scale-pretrained supporting benchmark; these models differ in pretraining objectives and datasets. Mechanistic analysis leveraged an attention dilution index (ADI), an entropy-based metric quantifying attention diverted toward uninformative padding tokens, and centered kernel alignment (CKA) to measure feature reuse during MRI adaptation. ADI increased with zero-padding, while high feature reuse did not necessarily translate to improved downstream accuracy. To mitigate these issues, we introduced two interventions: a tumor-aware augmentation strategy to expand tumor appearance heterogeneity coverage, and an anisotropic cropping strategy to restore token efficiency. Fine-tuning with these strategies on identical rectal MRI datasets yielded detection rates of 91.1% (225/247) and 88.7% (219/247) for the primary SMIT and Swin UNETR backbones, with the supporting VoCo benchmark reaching 90.3% (223/247), demonstrating significantly improved robustness under CT-to-MRI transfer. This study is among the first to examine when pretrained transformers fail to transfer across imaging modalities and demonstrates how targeted mitigation strategies can systematically overcome cross-modality transfer limitations.

Unlike vision and language domains, graph learning lacks a shared input space, as input features differ across graph datasets not only in semantics, but also in value ranges and dimensionality. This misalignment prevents graph models from generalizing across datasets, limiting their use as foundation models. In this work, we propose ALL-IN, a simple and theoretically grounded method that enables transferability across datasets with different input features. Our approach projects node features into a shared random space and constructs representations via covariance-based statistics, thus eliminating dependence on the original feature space. We show that the computed node-covariance operators and the resulting node representations are invariant in distribution to permutations of the input features. We further demonstrate that the expected operator exhibits invariance to general orthogonal transformations of the input features. Empirically, ALL-IN achieves strong performance across diverse node- and graph-level tasks on unseen datasets with new input features, without requiring architecture changes or retraining. These results point to a promising direction for input-agnostic, transferable graph models.

Deep learning methods are widely used under diverse resource constraints, resulting in models of varying sizes, such as the Vision Transformer (ViT) series. Deploying these models typically requires costly pretraining and finetuning. The Learngene paradigm addresses this issue by extracting transferable components, called learngenes, from a pretrained ancestry model (Ans-Net) to initialize variable-sized descendant models (Des-Nets).Existing learngene extraction methods rely on a single dataset, limiting downstream performance. To address this limitation, we propose Learngene Search Across Multiple Datasets for Building Variable-Sized Models (LSAMD). LSAMD expands the Ans-Net into a searchable super Ans-Net with dataset-specific blocks and dataset adapters (DADs). During training, LSAMD searches for an optimal architecture path for each dataset. The base blocks most frequently selected across datasets are extracted as learngenes for initializing Des-Nets.Experiments on multiple datasets show that LSAMD achieves performance comparable to pretrain-finetune methods while significantly reducing storage and training costs.

The increasing availability of advanced image editing tools has led to a significant rise in manipulated digital content, posing serious challenges for digital forensics and information security. This study presents a transfer learning-based framework for digital image forgery detection that integrates compression-aware feature enhancement with deep convolutional neural network (CNN) architectures. The proposed approach introduces a hybrid input representation that combines RGB images with compression difference-based features (FDIFF), explicitly highlighting subtle manipulation artifacts that are often difficult to detect. In addition, a model-specific adaptive threshold optimization strategy based on the Youden Index is employed to improve classification reliability by achieving a better balance between true positive and false positive rates. Experiments conducted on the CASIA v2.0 dataset using multiple pretrained CNN architectures, including DenseNet121, VGG16, ResNet50, EfficientNetB0, MobileNet, and InceptionV3, demonstrate the effectiveness and robustness of the proposed framework. The models are evaluated using comprehensive performance metrics such as accuracy, precision, recall, F1-score, Matthews correlation coefficient (MCC), and area under the ROC curve (AUC). The results show that DenseNet121 achieves the highest accuracy and AUC, while ResNet50 provides the most balanced and reliable predictions with the highest MCC. The findings emphasize that relying solely on accuracy is insufficient for forensic applications, where minimizing false negatives is critical. Overall, the proposed framework improves the visibility of manipulation artifacts and enhances classification robustness, making it suitable for real-world digital image forgery detection scenarios.

Automated sleep stage classification typically employs a single population-agnostic model, disregarding established demographic variations in sleep architecture. Sleep patterns, however, differ substantially across gender, age, and obstructive sleep apnea (OSA) severity, indicating that a onesize-fits all approach may be suboptimal for diverse clinical populations. In this paper, we propose a two stage training strategy based on demographic stratification and transfer learning framework. We first pretrains a convolutional recurrent model on the full population and then fine tunes it independently for demographic subgroups defined by gender, age, and Apnea-Hypopnea Index (AHI) severity according to the AASM clinical standard. Using the DREAMT dataset comprising 100 clinical subjects and 7 PSG channels, we evaluate 37 fine-tuned configurations across single-axis and two-way demographic combinations. Results demonstrate that 35 of the 37 fine-tuned models outperform the baseline, with Cohen's kappa improvements ranging from 0.9 to 12.9%. These findings indicate that stratified fine tuning tailored to specific patient demographics yields substantially more accurate sleep staging than a single generalized model, offering a practical and clinically grounded paradigm for personalized sleep assessment.

Personalized models are essential in digital health because individuals exhibit substantial physiological and behavioral heterogeneity. Yet personalization is limited by scarce and noisy user-specific data. Most existing methods rely on population pretraining or data from similar users only, which can lead to biased transfer and weak generalization. We propose a unified personalization framework that trains a personal model using adaptively weighted support users, including both similar and dissimilar individuals. The objective integrates personal loss, similarity-weighted transfer from similar users, and contrastive regularization from dissimilar users to suppress misleading correlations. An iterative optimization algorithm jointly updates model parameters and user similarity weights. Experiments on six tasks across four real-world digital health datasets show consistent improvements over population and personalized baselines. The method achieves up to 10% lower RMSE on large-scale datasets and approximately 25% lower RMSE in low-data settings. The learned adaptive weights improve data efficiency and provide interpretable guidance for targeted data selection.

Physics-Informed Neural Networks (PINNs) offer a flexible paradigm for solving differential equations by embedding governing laws into the training objective. A persistent limitation is instance specificity: standard PINNs typically require retraining for each new forcing term, boundary/initial condition, or parameter setting. One-shot transfer learning (OTL) addresses this bottleneck for linear operators by freezing a pretrained latent representation and computing optimal output weights in closed form, but for nonlinear problems closed-form adaptation is generally unavailable because the loss is nonconvex in the output layer. In this paper we substantially broaden the class of nonlinearities amenable to one-shot PINN transfer by combining OTL with Chebyshev polynomial surrogates. We approximate general smooth weakly nonlinear terms by truncated Chebyshev expansions over a prescribed solution range, yielding a polynomial nonlinearity that can be handled by a perturbative decomposition into linear subproblems. A multi-head PINN learns a reusable latent space associated with the dominant linear operator; at test time, solutions to new instances are obtained via a sequence of closed-form linear solves in the output layer, without retraining the network body. We provide a unified derivation of the framework for ODEs and PDEs and demonstrate accuracy and fast online adaptation on nonlinear benchmarks, including non-polynomial and singular ODE nonlinearities as well as a reaction-diffusion PDE with saturating kinetics, demonstrating the method's utility in many-query regimes.

Robust control policy learning for autonomous driving requires training environments to be both physically realistic and computationally scalable, properties that existing simulators provide only in isolation. We introduce Sim2Sim2Sim, a framework that bridges high-fidelity vehicle simulation and scalable reinforcement learning by distilling simulator dynamics into a highly parallelizable learned dynamics model. By training control policies purely within this distilled environment and deploying them back into the high-fidelity source simulator, we demonstrate more efficient policy optimization and reliable transfer under challenging dynamics. We further show that predictive accuracy alone does not fully characterize a learned dynamics model's suitability as a reinforcement learning training environment, which should also be assessed by the quality of the policies it enables.

Plants, crops and their yields are essential to our very existence, but diseases and pests cause large losses every year. As such it is vital to ensure that diseases can be spotted early and treated accordingly and stopping the spread while still possible. Manual and traditional methods require personal to walk through the field and check for symptoms 'by hand'. This is very laborious and very time consuming, so ML methods have been applied as a result and they have garnered promising results. CNN models are especially efficient as they can automatically extract features from images without any manual feature construction before then feeding the features to a classifier. Datasets are largely influential to the final performance of the model. Despite the importance that datasets pose to the field, there still seems to be somewhat of a discrepancy between what is publicly available for use and what would be required to sufficiently train fully capable models. To overcome these shortcomings, as part of this thesis open datasets for the field of plant leaf disease classification have been identified as well as models that can be trained on them and extensive benchmarks have been carried out to identify their suitability. Then a new dataset was constructed based on those findings as well as on the findings of a augmentation applicability study, which will be used to train a new Base Model based on the DenseNet201 architecture, which managed to outperform the baseline model on said new dataset as well as outperforming it on plant leaf disease classification domain specific Transfer-Learning experiments on another new dataset. This new model manages to train models through Transfer-Learning (TL) faster, more robust, more stable, and with less data than general model would, overcoming a large number of issues that the field still suffers from.

Frozen Gemma 4 31B weights pretrained exclusively on text, unmodified, transfer through a thin trainable interface to non-text modalities the substrate has never processed. On the L24--L29 slice (192 attention heads), an English-text TxtCopy attention probe (95 sentences) and per-head ablation impact on four non-language token-pattern tasks (binary copy, associative recall, 1D cellular automaton Rule 90, binary addition) jointly classify four heads -- L26.28, L27.28, L27.2, L27.3 -- as top-tier on both signals. The slice-level joint coincidence is significant under hypergeometric null ($P = 0.0013$, $N=192$, $K=38$, $n=4$) and survives multiplicity-aware permutation tests ($P_{V4} = 0.013$). Pretrained Gemma L26 reaches 60.22% on OGBench cube-double-play-task1 vs ~1% for random-init Gemma ($+59$pt at $n=3$); a FrozenRandom-GPT2 control with correct $1/\sqrt{d_k}$ scaling also fails. Head-level causal validation: zeroing L26.28 in the trained cube-task1 IQL agent drops success $63.3\% \to 10.0\%$ vs $46.7\%$ for a layer-matched low-TxtCopy negative control ($3.2\times$ specificity at $n=30$; $n=5$ paired-$t$ $p=0.039$). A full L26 sweep places L26.28 at rank 4 of 32. Honest negatives: within-L26 Spearman $ρ(\text{TxtCopy, drop}) = +0.37$ (opposite of within-layer causal reading); single-head activation patching does not transfer the matching variable; the 4 named heads alone do not suffice on any task; Walker2d-DT and scene-task1 recruit L24 outside the named slice and show null head-ablation specificity. We frame the contribution as a cross-distribution importance fingerprint at the slice level plus head-level causal evidence on one cross-modality target.

Memory-augmented LLM agents offer an appealing shortcut to continual learning: rather than updating model parameters, they accumulate experience in external memory, seemingly sidestepping the stability-plasticity dilemma of parametric learning. We show that this challenge does not disappear but resurfaces at the memory level. Under a limited context window, old and new experiences compete during retrieval, relocating the continual-learning bottleneck from parameter updates to memory access. To study this phenomenon, we introduce a (k,v) framework that disentangles two fundamental design axes of external memory: how experience is represented and how it is organized for retrieval. Across sequential-task experiments in ALFWorld and BabyAI, we find that abstract procedural memories transfer more reliably than detailed trajectories, while negative transfer disproportionately harms the hard cases. Moreover, finer-grained memory organization is not universally beneficial: designs that yield strong forward transfer can simultaneously induce severe forgetting. Together, these results reveal that external memory does not resolve the continual-learning problem; it reshapes it into a problem of memory representation and retrieval design.