arXiv AI & Jobs Research (Lorena A. Barba, Laura Stegner) • Published: 2026-01-15
Traditional assessment methods collapse when students use generative AI to complete work without genuine engagement, creating an illusion of competence where they believe they're learning but aren't. This paper presents the conversational exam -- a scalable oral examination format that restores assessment validity by having students code live while
arXiv AI & Jobs Research (Maissam Barkeshli, Alberto Alfarano, Andrey Gromov) • Published: 2026-01-15
Scaling laws have played a major role in the modern AI revolution, providing practitioners predictive power over how the model performance will improve with increasing data, compute, and number of model parameters. This has spurred an intense interest in the origin of neural scaling laws, with a common suggestion being that they arise from power la
arXiv AI & Jobs Research (Laura Ferrarotti, Gian Maria Campedelli, Roberto D) • Published: 2026-01-15
In this article, we argue that understanding the collective behavior of agents based on large language models (LLMs) is an essential area of inquiry, with important implications in terms of risks and benefits, impacting us as a society at many levels. We claim that the distinctive nature of LLMs--namely, their initialization with extensive pre-trai
arXiv AI & Jobs Research (Reza M. Asiyabi, SEOSAW Partnership, Steven Hancoc) • Published: 2026-01-15
Concept Bottleneck Models (CBMs) improve the explainability of black-box Deep Learning (DL) by introducing intermediate semantic concepts. However, standard CBMs often overlook domain-specific relationships and causal mechanisms, and their dependence on complete concept labels limits applicability in scientific domains where supervision is sparse b
arXiv AI & Jobs Research (Frank Bobe, Gregory D. Vetaw, Chase Pavlick, Darsh) • Published: 2026-01-15
The practice of fine-tuning Large Language Models (LLMs) has achieved state-of-the-art performance on specialized tasks, yet diagnosing why these models become brittle and fail to generalize remains a critical open problem. To address this, we introduce and apply a multi-layered diagnostic framework to a cross-architectural study. We fine-tune Llam
arXiv AI & Jobs Research (Daniyaal Farooqi, Gavin Pu, Shreyasha Paudel, Shar) • Published: 2026-01-15
The emerging widespread usage of AI has led to industry adoption to improve efficiency and increase earnings. However, a major consequence of this is AI displacing employees from their jobs, leading to feelings of job insecurity and uncertainty. This is especially true for computer science students preparing to enter the workforce. To investigate t
AI Skill Now