Low Resource Style Transfer via Domain Adaptive Meta Learning

Domain generalizable (DG) person re-identification (ReID) aims to test across unseen domains without access to the target domain data at training time, which is a realistic but challenging problem. In contrast to methods assuming an identical model for different domains, Mixture of Experts (MoE) exploits multiple domain-specific networks for leveraging complementary information between domains, obtaining impressive results. However, prior MoE-based DG ReID methods suffer from a large model size with the increase of the number of source domains, and most of them overlook the exploitation of domain-invariant characteristics. To handle the two issues above, this paper presents a new approach called Mimic Embedding via adapTive Aggregation (\(\textsf{META}\)) for DG person ReID. To avoid the large model size, experts in \(\textsf{META}\) do not adopt a branch network for each source domain but share all the parameters except for the batch normalization layers. Besides multiple experts, \(\textsf{META}\) leverages Instance Normalization (IN) and introduces it into a global branch to pursue invariant features across domains. Meanwhile, \(\textsf{META}\) considers the relevance of an unseen target sample and source domains via normalization statistics and develops an aggregation module to adaptively integrate multiple experts for mimicking unseen target domain. Benefiting from a proposed consistency loss and an episodic training algorithm, \(\textsf{META}\) is expected to mimic embedding for a truly unseen target domain. Extensive experiments verify that \(\textsf{META}\) surpasses state-of-the-art DG person ReID methods by a large margin. Our code is available at https://github.com/xbq1994/META. KeywordsDomain generalizationPerson re-identification

The style transfer task (here style is used in a broad “authorial” sense with many aspects including register, sentence structure, and vocabulary choice) takes text input and rewrites it in a specified target style preserving the meaning, but altering the style of the source text to match that of the target. Much of the existing research on this task depends on the use of parallel datasets. In this work we employ recent results in unsupervised cross-lingual language modeling (XLM) and machine translation to effect style transfer while treating the input data as unaligned. First, we show that adding “content embeddings” to the XLM which capture human-specified groupings of subject matter can improve performance over the baseline model. Evaluation of style transfer has often relied on metrics designed for machine translation which have received criticism of their suitability for this task. As a second contribution, we propose the use of a suite of classical stylometrics as a useful complement for evaluation. We select a few such measures and include these in the analysis of our results.

Unsupervised text style transfer aims to alter text styles while preserving the content, without aligned data for supervision. Existing seq2seq methods face three challenges: 1) the transfer is weakly interpretable, 2) generated outputs struggle in content preservation, and 3) the trade-off between content and style is intractable. To address these challenges, we propose a hierarchical reinforced sequence operation method, named Point-Then-Operate (PTO), which consists of a high-level agent that proposes operation positions and a low-level agent that alters the sentence. We provide comprehensive training objectives to control the fluency, style, and content of the outputs and a mask-based inference algorithm that allows for multi-step revision based on the single-step trained agents. Experimental results on two text style transfer datasets show that our method significantly outperforms recent methods and effectively addresses the aforementioned challenges.

Artistic text style transfer is the task of migrating the style from a source image to the target text to create artistic typography. Recent style transfer methods have considered texture control to enhance usability. However, controlling the stylistic degree in terms of shape deformation remains an important open challenge. In this paper, we present the first text style transfer network that allows for real-time control of the crucial stylistic degree of the glyph through an adjustable parameter. Our key contribution is a novel bidirectional shape matching framework to establish an effective glyph-style mapping at various deformation levels without paired ground truth. Based on this idea, we propose a scale-controllable module to empower a single network to continuously characterize the multi-scale shape features of the style image and transfer these features to the target text. The proposed method demonstrates its superiority over previous state-of-the-arts in generating diverse, controllable and high-quality stylized text.

Ensuring that deep learning models generalize across distinct regimes remains a fundamental challenge in Earth system modeling. Due to the inherent violation of the independent and identically distributed (i.i.d.) assumption, models optimized for local conditions rarely exhibit robust performance on unseen domains. While Unsupervised Domain Adaptation (UDA) is a well-established technique for mitigating such distribution shifts in computer vision, its application to Earth system modeling remains underexplored. In this study we investigate the efficacy of UDA for the super-resolution of atmospheric fields, utilizing kilometer-scale COSMO simulations [1] and the RainShift benchmark dataset [2] to quantify model robustness across different regions. We apply residual learning to jointly super-resolve precipitation and surface pressure, incorporating static predictors such as topography. To quantify transferability, we propose a systematic framework that trains on source domains and evaluates on unseen target domains, treating spatial transfer as a proxy for model robustness under distribution shifts. We introduce a consistency metric to evaluate model adaptation by comparing mean performance on seen versus unseen domains. We assess a hierarchy of adaptation methods, ranging from simple regularization to physics-informed approaches. These include domain-specific regularization and distribution alignment methods, domain adversarial training, and geometry-robust training via group-equivariant convolutions. Preliminary results on the COSMO simulations demonstrate that even elementary adaptation strategies, such as dropout and data augmentation, improve cross-domain consistency. This work establishes a controlled setup for benchmarking generalization, suggesting that UDA offers a viable pathway to bridge the gap between locally trained models and global applicability.[1]: Cui, R., Thurnherr, I., Velasquez, P., Brennan, K. P., Leclair, M., Mazzoleni, A., et al. (2025). A European hail and lightning climatology from an 11-year kilometer-scale regional climate simulation. Journal of Geophysical Research: Atmospheres, 130, e2024JD042828. https://doi.org/10.1029/2024JD042828[2]: Paula Harder et al. RainShift: A Benchmark for Precipitation Downscaling Across Geographies. 2025. arXiv: 2507.04930 [cs.CV]. url: https://arxiv.org/abs/2507.04930.

Domain Generalization (DG) aims to generalize a model trained on source domains to unseen target domains. Learning domain-invariant representations based on causal inference is one of the popular directions in DG. However, these methods would yield an inaccurate causal variable set due to the lack of heterogeneous domain data or a prior causal structure, which severely weakens their generalization capacity. To this end, we propose a novel DG method called Perturbing Confounders via Causal Disentanglement (PCCD), which explicitly disentangles latent features into causally relevant features and confounding features. The method perturbs the confounding features to improve generalization on unseen domains. Specifically, we first apply the causal disentanglement framework to separate causal features and confounding features. Then, we introduce a learnable perturbation initialized as Gaussian distribution on the batch-wise statistics for each dimension of the confounding features while constraining semantic consistency. Extensive experiments on several benchmarks indicate that our framework achieves state-of-the-art performance compared to other competing methods.

Dynamic domain generalization for medical image segmentation.

We introduce a simple method for text style transfer that frames style transfer as denoising: we synthesize a noisy corpus and treat the source style as a noisy version of the target style. To control for aspects such as preserving meaning while modifying style, we propose a reranking approach in the data synthesis phase. We evaluate our method on three novel style transfer tasks: transferring between British and American varieties, text genres (formal vs. casual), and lyrics from different musical genres. By measuring style transfer quality, meaning preservation, and the fluency of generated outputs, we demonstrate that our method is able both to produce high-quality output while maintaining the flexibility to suggest syntactically rich stylistic edits.

In recent years, researchers in computational linguistics have shown a growing interest in the style of text, with a specific focus on the text style transfer (TST) task. While numerous innovative methods have been proposed, it has been observed that the existing evaluations are insufficient to validate the claims and precisely measure the performance. This challenge primarily stems from rapid advancements and diverse settings of these methods, with the associated (re)implementation and reproducibility hurdles. To bridge this gap, we introduce a comprehensive benchmark for TST known as TSTBench. TSTBench includes a codebase encompassing implementations of 13 state-of-the-art algorithms and a standardized protocol for text style transfer. Based on the codebase and protocol, we have conducted thorough experiments across seven datasets, resulting in a total of 7000+ evaluations. Our work provides extensive analysis from various perspectives, explores the performance of representative baselines across various datasets, and offers insights into the task and evaluation processes to guide future research in TST.

GradCa: Generalizing to unseen domains via gradient calibration

Recent advances in pre-trained language models (PLMs), such as GPT, BERT, and their larger variants, have significantly enhanced the ability to generate synthetic datasets for zero-shot learning, where models can generate data without the need for human annotations. However, existing methods often focus on generating domain-specific datasets, which restricts their generalization capabilities when applied to new, unseen domains. Furthermore, many of these methods rely on a single, static dataset that is either too broad or too narrowly tailored to a specific domain, making them less adaptable to diverse downstream domains. In this paper, we propose a dual prompt data generation framework, called Adaptive Domain-Specific Classification via Zero-shot Dataset Generation (AdaGen), which decouples universal representation learning from domain-specific adaptation. The framework consists of two components: domainagnostic training set generation via a universal prompt, followed by domain-specific validation set generation for weight learning. By combining these two components, our method achieves an optimal trade-off between cross-domain robustness and domainspecific accuracy, allowing a small task-aware model to effectively adapt across diverse domains while preserving generalization. This novel method overcomes the limitations of traditional methods by introducing dynamic feedback, enabling efficient zeroshot learning with improved adaptability. Extensive experiments in seven real-world datasets demonstrate the effectiveness of AdaGen, which outperforms the compared baseline models.

Text style transfer (TST) involves altering the linguistic style of a text while preserving its style-independent content.This paper focuses on sentiment transfer, a popular TST subtask, across a spectrum of Indian languages: Hindi, Magahi, Malayalam, Marathi, Punjabi, Odia, Telugu, and Urdu, expanding upon previous work on English-Bangla sentiment transfer (Mukherjee et al., 2023a).We introduce dedicated datasets of 1,000 positive and 1,000 negative style-parallel sentences for each of these eight languages.We then evaluate the performance of various benchmark models categorized into parallel, non-parallel, cross-lingual, and shared learning approaches, including the Llama2 and GPT-3.5 large language models (LLMs).Our experiments highlight the significance of parallel data in TST and demonstrate the effectiveness of the Masked Style Filling (MSF) approach (Mukherjee et al., 2023a) in non-parallel techniques.Moreover, crosslingual and joint multilingual learning methods show promise, offering insights into selecting optimal models tailored to the specific language and task requirements.To the best of our knowledge, this work represents the first comprehensive exploration of the TST task as sentiment transfer across a diverse set of languages.

Natural language generation (NLG) spans a broad range of tasks, each of which serves for specific objectives and desires different properties of generated text. The complexity makes automatic evaluation of NLG particularly challenging. Previous work has typically focused on a single task and developed individual evaluation metrics based on specific intuitions. In this paper, we propose a unifying perspective based on the nature of information change in NLG tasks, including compression (e.g., summarization), transduction (e.g., text rewriting), and creation (e.g., dialog). Information alignment between input, context, and output text plays a common central role in characterizing the generation. With automatic alignment prediction models, we develop a family of interpretable metrics that are suitable for evaluating key aspects of different NLG tasks, often without need of gold reference data. Experiments show the uniformly designed metrics achieve stronger or comparable correlations with human judgement compared to state-of-the-art metrics in each of diverse tasks, including text summarization, style transfer, and knowledge-grounded dialog.

Existing text style transfer (TST) methods rely on style classifiers to disentangle the text’s content and style attributes for text style transfer. While the style classifier plays a critical role in existing TST methods, there is no known investigation on its effect on the TST methods. In this paper, we conduct an empirical study on the limitations of the style classifiers used in existing TST methods. We demonstrated that the existing style classifiers cannot learn sentence syntax effectively and ultimately worsen existing TST models’ performance. To address this issue, we propose a novel Syntax-Aware Controllable Generation (SACG) model, which includes a syntax-aware style classifier that ensures learned style latent representations effectively capture the sentence structure for TST. Through extensive experiments on two popular text style transfer tasks, we show that our proposed method significantly outperforms twelve state-of-the-art methods. Our case studies have also demonstrated SACG’s ability to generate fluent target-style sentences that preserved the original content.

Text style transfer is an important issue for conversational agents as it may adapt utterance production to specific dialogue situations. It consists in introducing a given style within a sentence while preserving its semantics. Within this scope, different strategies have been proposed that either rely on parallel data or take advantage of non-supervised techniques. In this paper, we follow the latter approach and show that the sequential introduction of different loss functions into the learning process can boost the performance of a standard model. We also evidence that combining different style classifiers that either focus on global or local textual information improves sentence generation. Experiments on the Yelp dataset show that our methodology strongly competes with the current state-of-the-art models across style accuracy, grammatical correctness, and content preservation.