Bridging linguistic gaps fosters global growth and cultural exchange. This study addresses the challenges of Roman Urdu -- a Latin-script adaptation of Urdu widely used in digital communication -- by creating a novel parallel dataset comprising 75,146 sentence pairs. Roman Urdu's lack of standardization, phonetic variability, and code-switching with English complicates language processing. We tackled this by employing a hybrid approach that combines synthetic data generated via advanced prompt engineering with real-world conversational data from personal messaging groups. We further refined the dataset through a human evaluation phase, addressing linguistic inconsistencies and ensuring accuracy in code-switching, phonetic representations, and synonym variability. The resulting dataset captures Roman Urdu's diverse linguistic features and serves as a critical resource for machine translation, sentiment analysis, and multilingual education.

The availability of multiple training algorithms and architectures for generative models requires a selection mechanism to form a single model over a group of well-trained generation models. The selection task is commonly addressed by identifying the model that maximizes an evaluation score based on the diversity and quality of the generated data. However, such a best-model identification approach overlooks the possibility that a mixture of available models can outperform each individual model. In this work, we explore the selection of a mixture of multiple generative models and formulate a quadratic optimization problem to find an optimal mixture model achieving the maximum of kernel-based evaluation scores including kernel inception distance (KID) and R\'{e}nyi kernel entropy (RKE). To identify the optimal mixture of the models using the fewest possible sample queries, we propose an online learning approach called Mixture Upper Confidence Bound (Mixture-UCB). Specifically, our proposed online learning method can be extended to every convex quadratic function of the mixture weights, for which we prove a concentration bound to enable the application of the UCB approach. We prove a regret bound for the proposed Mixture-UCB algorithm and perform several numerical experiments to show the success of the proposed Mixture-UCB method in finding the optimal mixture of text-based and image-based generative models. The codebase is available at https://github.com/Rezaei-Parham/Mixture-UCB .

Document similarity assessment plays an important role in various natural language processing (NLP) applications, such as information retrieval, plagiarism detection, recommendation systems, and question answering [11, 19]. For instance, in recommendation systems, document similarity helps personalise suggestions by finding content that closely matches user preference. These tasks rely on accurate measurements of how similar documents are in terms of their structure, content, and meaning, which depends on the way the document is represented computationally. This representation is usually done in vector format and is obtained via document embedding methods. V arious methodologies can be employed to obtain document-level embeddings, and the choice of method directly impacts the accuracy and usefulness of the similarity scores calculated [14, 19].

As foundation models increasingly permeate sensitive domains such as healthcare, finance, and mental health, ensuring their behavior meets desired outcomes and social expectations becomes critical. Given the complexities of these high-dimensional models, traditional techniques for constraining agent behavior, which typically rely on low-dimensional, discrete state and action spaces, cannot be directly applied. Drawing inspiration from robotic action selection techniques, we propose the grounded observer framework for constraining foundation model behavior that offers both behavioral guarantees and real-time variability. This method leverages real-time assessment of low-level behavioral characteristics to dynamically adjust model actions and provide contextual feedback. To demonstrate this, we develop a system capable of sustaining contextually appropriate, casual conversations ("small talk"), which we then apply to a robot for novel, unscripted interactions with humans. Finally, we discuss potential applications of the framework for other social contexts and areas for further research.

In this paper, we study a system in which a sensor forwards status updates to a receiver through an error-prone channel, while the receiver sends the transmission results back to the sensor via a reliable channel. Both channels are subject to random delays. To evaluate the timeliness of the status information at the receiver, we use the Age of Information (AoI) metric. The objective is to design a sampling policy that minimizes the expected time-average AoI, even when the channel statistics (e.g., delay distributions) are unknown. We first review the threshold structure of the optimal offline policy under known channel statistics and then reformulate the design of the online algorithm as a stochastic approximation problem. We propose a Robbins-Monro algorithm to solve this problem and demonstrate that the optimal threshold can be approximated almost surely. Moreover, we prove that the cumulative AoI regret of the online algorithm increases with rate $\mathcal{O}(\ln K)$, where $K$ is the number of successful transmissions. In addition, our algorithm is shown to be minimax order optimal, in the sense that for any online learning algorithm, the cumulative AoI regret up to the $K$-th successful transmissions grows with the rate at least $\Omega(\ln K)$ in the worst case delay distribution. Finally, we improve the stability of the proposed online learning algorithm through a momentum-based stochastic gradient descent algorithm. Simulation results validate the performance of our proposed algorithm.

The item cold-start problem is crucial for online recommender systems, as the success of the cold-start phase determines whether items can transition into popular ones. Prompt learning, a powerful technique used in natural language processing (NLP) to address zero- or few-shot problems, has been adapted for recommender systems to tackle similar challenges. However, existing methods typically rely on content-based properties or text descriptions for prompting, which we argue may be suboptimal for cold-start recommendations due to 1) semantic gaps with recommender tasks, 2) model bias caused by warm-up items contribute most of the positive feedback to the model, which is the core of the cold-start problem that hinders the recommender quality on cold-start items. We propose to leverage high-value positive feedback, termed pinnacle feedback as prompt information, to simultaneously resolve the above two problems. We experimentally prove that compared to the content description proposed in existing works, the positive feedback is more suitable to serve as prompt information by bridging the semantic gaps. Besides, we propose item-wise personalized prompt networks to encode pinnaclce feedback to relieve the model bias by the positive feedback dominance problem. Extensive experiments on four real-world datasets demonstrate the superiority of our model over state-of-the-art methods. Moreover, PROMO has been successfully deployed on a popular short-video sharing platform, a billion-user scale commercial short-video application, achieving remarkable performance gains across various commercial metrics within cold-start scenarios

In the domain of second-language acquisition, predictive modeling serves as a pivotal tool for facilitating educators in implementing diversified teaching strategies, thereby garnering extensive research attention. Despite the prevalent focus on model accuracy in most existing studies, the exploration into model fairness remains substantially underexplored. Model fairness pertains to the equitable treatment of different groups by machine learning algorithms. It ensures that the model's predictions do not exhibit unintentional biases against certain groups based on attributes such as gender, ethnicity, age, or other potentially sensitive characteristics. In essence, a fair model should produce outcomes that are impartial and do not perpetuate existing prejudices, ensuring that no group is systematically disadvantaged. In this research, we evaluate the fairness of two predictive models based on second-language learning, utilizing three tracks from the Duolingo dataset: en_es (English learners who speak Spanish), es_en(Spanish learners who speak English), and fr_en(French learners who speak English). We measure (i) algorithmic fairness among different clients such as iOS, Android and Web and (ii) algorithmic fairness between developed countries and developing countries. Our findings indicate: 1) Deep learning exhibits a marked advantage over machine learning when applied to knowledge tracing based on second language acquisition, owing to its heightened accuracy and fairness.

We introduce the first highly multilingual speech and American Sign Language (ASL) comprehension dataset by extending BELEBELE. Our dataset covers 74 spoken languages at the intersection of BELEBELE and FLEURS, and one sign language (ASL). We evaluate 2M-BELEBELE dataset for both 5-shot and zero-shot settings and across languages, the speech comprehension accuracy is ~ 2-3% average lower compared to reading comprehension.

Few shot and Chain-of-Thought prompting have shown promise when applied to Physics Question Answering Tasks, but are limited by the lack of mathematical ability inherent to LLMs, and are prone to hallucination. By utilizing a Mixture of Experts (MoE) Model, along with analogical prompting, we are able to show improved model performance when compared to the baseline on standard LLMs. We also survey the limits of these prompting techniques and the effects they have on model performance. Additionally, we propose Analogical CoT prompting, a prompting technique designed to allow smaller, open source models to leverage Analogical prompting, something they have struggled with, possibly due to a lack of specialist training data.

Diffusion models have emerged as powerful tools for generative modeling, demonstrating exceptional capability in capturing target data distributions from large datasets. However, fine-tuning these massive models for specific downstream tasks, constraints, and human preferences remains a critical challenge. While recent advances have leveraged reinforcement learning algorithms to tackle this problem, much of the progress has been empirical, with limited theoretical understanding. To bridge this gap, we propose a stochastic control framework for fine-tuning diffusion models. Building on denoising diffusion probabilistic models as the pre-trained reference dynamics, our approach integrates linear dynamics control with Kullback-Leibler regularization. We establish the well-posedness and regularity of the stochastic control problem and develop a policy iteration algorithm (PI-FT) for numerical solution. We show that PI-FT achieves global convergence at a linear rate. Unlike existing work that assumes regularities throughout training, we prove that the control and value sequences generated by the algorithm maintain the regularity. Additionally, we explore extensions of our framework to parametric settings and continuous-time formulations.