Generative AI and large language models hold great promise in enhancing programming education by generating individualized feedback and hints for learners. Recent works have primarily focused on improving the quality of generated feedback to achieve human tutors' quality.

Generative AI and large language models hold great promise in enhancing programming education by generating individualized feedback and hints for learners. Recent works have primarily focused on improving the quality of generated feedback to achieve human tutors' quality.

This is the first work to investigate the effectiveness of BERT-based contextual embeddings in active learning (AL) tasks on cold-start scenarios, where traditional fine-tuning is infeasible due to the absence of labeled data. Our primary contribution is the proposal of a more robust fine-tuning pipeline - DoTCAL - that diminishes the reliance on labeled data in AL using two steps: (1) fully leveraging unlabeled data through domain adaptation of the embeddings via masked language modeling and (2) further adjusting model weights using labeled data selected by AL. Our evaluation contrasts BERT-based embeddings with other prevalent text representation paradigms, including Bag of Words (BoW), Latent Semantic Indexing (LSI), and FastText, at two critical stages of the AL process: instance selection and classification. Experiments conducted on eight ATC benchmarks with varying AL budgets (number of labeled instances) and number of instances (about 5,000 to 300,000) demonstrate DoTCAL's superior effectiveness, achieving up to a 33% improvement in Macro-F1 while reducing labeling efforts by half compared to the traditional one-step method. We also found that in several tasks, BoW and LSI (due to information aggregation) produce results superior (up to 59% ) to BERT, especially in low-budget scenarios and hard-to-classify tasks, which is quite surprising.

Generative AI and large language models hold great promise in enhancing programming education by generating individualized feedback and hints for learners. Recent works have primarily focused on improving the quality of generated feedback to achieve human tutors' quality. While quality is an important performance criterion, it is not the only criterion to optimize for real-world educational deployments. In this paper, we benchmark language models for programming feedback generation across several performance criteria, including quality, cost, time, and data privacy. The key idea is to leverage recent advances in the new paradigm of in-browser inference that allow running these models directly in the browser, thereby providing direct benefits across cost and data privacy. To boost the feedback quality of small models compatible with in-browser inference engines, we develop a fine-tuning pipeline based on GPT-4 generated synthetic data. We showcase the efficacy of fine-tuned Llama3-8B and Phi3-3.8B 4-bit quantized models using WebLLM's in-browser inference engine on three different Python programming datasets. We will release the full implementation along with a web app and datasets to facilitate further research on in-browser language models.

Large Language Models (LLMs) exploit fine-tuning as a technique to adapt to diverse goals, thanks to task-specific training data. Task specificity should go hand in hand with domain orientation, that is, the specialization of an LLM to accurately address the tasks of a given realm of interest. However, models are usually fine-tuned over publicly available data or, at most, over ground data from databases, ignoring business-level definitions and domain experience. On the other hand, Enterprise Knowledge Graphs (EKGs) are able to capture and augment such domain knowledge via ontological reasoning. With the goal of combining LLM flexibility with the domain orientation of EKGs, we propose a novel neurosymbolic architecture that leverages the power of ontological reasoning to build task- and domain-specific corpora for LLM fine-tuning.