January 2026

arXiv:2601.09714v1 Announce Type: new Abstract: The integration of Large Language Models (LLMs) into the scientific ecosystem raises fundamental questions about the creativity and originality of AI-generated research. Recent work has identified “smart plagiarism” as a concern in single-step prompting approaches, where models reproduce existing ideas with terminological shifts. This paper investigates whether agentic workflows — multi-step systems employing iterative reasoning, evolutionary search, and recursive decomposition — can generate more novel and feasible research plans. We benchmark five reasoning […]

Ver mais

arXiv:2601.09713v1 Announce Type: new Abstract: Proactively predicting a users next utterance in human-machine dialogue can streamline interaction and improve user experience. Existing commercial API-based solutions are subject to privacy concerns while deploying general-purpose LLMs locally remains computationally expensive. As such, training a compact, task-specific LLM provides a practical alternative. Although user simulator methods can predict a user’s next utterance, they mainly imitate their speaking style rather than advancing the dialogue. Preference data synthesis has been investigated to generate […]

Ver mais

arXiv:2601.09712v1 Announce Type: new Abstract: Behavioral targeting, or online profiling, is a hotly debated topic. Much of the collection of personal information on the Internet is related to behavioral targeting, although research suggests that most people don’t want to receive behaviorally targeted advertising. The World Wide Web Consortium is discussing a Do Not Track standard, and regulators worldwide are struggling to come up with answers. This article discusses European law and recent policy developments on behavioral targeting.

Ver mais

arXiv:2601.09711v1 Announce Type: new Abstract: This paper discusses legal developments in Europe and the Netherlands. Recent decisions show that European data protection law, or privacy law, applies to behavioral targeting in most cases. Dutch law explicitly presumes that data protection law applies to behavioral targeting. This means that companies have to comply with data protection law’s fair information principles. For example, companies must refrain from secret or excessive data collection. Perhaps the principles could provide inspiration for future […]

Ver mais

arXiv:2601.09709v1 Announce Type: new Abstract: Social Determinants of Health correlate with patient outcomes but are rarely captured in structured data. Recent attention has been given to automatically extracting these markers from clinical text to supplement diagnostic systems with knowledge of patients’ social circumstances. Large language models demonstrate strong performance in identifying Social Determinants of Health labels from sentences. However, prediction in large admissions or longitudinal notes is challenging given long distance dependencies. In this paper, we explore hospital […]

Ver mais

I want to build a super fast rubiks cube simulator, I understand there is a math component on how to represent states & actions effectively, as well as, in a way that is compute efficient and fast, trying to look at some rotations and clean ways of representing it, but I do not have a computer architecture background, I want to get down, understand the basics of what operations make compute faster, and what’s more efficient, and how […]

Ver mais

Stochastic gradient descent (SGD) is central to deep learning, yet the dynamical origin of its preference for flatter, more generalizable solutions remains unclear. Here, by analyzing SGD learning dynamics, we identify a nonequilibrium mechanism governing solution selection. Numerical experiments reveal a transient exploratory phase in which SGD trajectories repeatedly escape sharp valleys and transition toward flatter regions of the loss landscape. By using a tractable physical model, we show that the SGD noise reshapes the landscape into an […]

Ver mais

Most pseudo-label selection strategies in semi-supervised learning rely on fixed confidence thresholds, implicitly assuming that prediction confidence reliably indicates correctness. In practice, deep networks are often overconfident: high-confidence predictions can still be wrong, while informative low-confidence samples near decision boundaries are discarded. This paper introduces a Confidence-Variance (CoVar) theory framework that provides a principled joint reliability criterion for pseudo-label selection. Starting from the entropy minimization principle, we derive a reliability measure that combines maximum confidence (MC) with residual-class […]

Ver mais

Published on January 2, 2026 4:29 AM GMT Past years: 2023 2024 Continuing a yearly tradition, I evaluate AI predictions from past years, and collect a convenience sample of AI predictions made this year. I prefer selecting specific predictions, especially ones made about the near term, enabling faster evaluation. Evaluated predictions made about 2025 in 2023, 2024, or 2025 mostly overestimate AI capabilities advances, although there’s of course a selection effect (people making notable predictions about the near-term are more […]

Ver mais

Metric-based few-shot approaches have gained significant popularity due to their relatively straightforward implementation, high interpret ability, and computational efficiency. However, stemming from the batch-independence assumption during testing, which prevents the model from leveraging valuable knowledge accumulated from previous batches. To address these challenges, we propose a novel test-time method called Incremental Prototype Enhancement Classifier (IPEC), a test-time method that optimizes prototype estimation by leveraging information from previous query samples. IPEC maintains a dynamic auxiliary set by selectively incorporating […]

Ver mais