A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness
arXiv:2603.06594v1 Announce Type: new Abstract: Automated \enquote{LLM-as-a-Judge} frameworks have become the de facto standard for scalable evaluation across natural language processing. For instance, in safety evaluation, these judges are relied upon to evaluate harmfulness in order to benchmark the robustness...
Taiwan Safety Benchmark and Breeze Guard: Toward Trustworthy AI for Taiwanese Mandarin
arXiv:2603.07286v1 Announce Type: new Abstract: Global safety models exhibit strong performance across widely used benchmarks, yet their training data rarely captures the cultural and linguistic nuances of Taiwanese Mandarin. This limitation results in systematic blind spots when interpreting region-specific risks...
How Much Noise Can BERT Handle? Insights from Multilingual Sentence Difficulty Detection
arXiv:2603.07346v1 Announce Type: new Abstract: Noisy training data can significantly degrade the performance of language-model-based classifiers, particularly in non-topical classification tasks. In this study we designed a methodological framework to assess the impact of denoising. More specifically, we explored a...
Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models
arXiv:2603.06621v1 Announce Type: new Abstract: Process Reward Models (PRMs) are rapidly becoming the backbone of LLM reasoning pipelines, yet we demonstrate that state-of-the-art PRMs are systematically exploitable under adversarial optimization pressure. To address this, we introduce a three-tiered diagnostic framework...
A new Uncertainty Principle in Machine Learning
arXiv:2603.06634v1 Announce Type: new Abstract: Many scientific problems in the context of machine learning can be reduced to the search of polynomial answers in appropriate variables. The Hevisidization of arbitrary polynomial is actually provided by one-and-the same two-layer expression. What...
Regression Models Meet Foundation Models: A Hybrid-AI Approach to Practical Electricity Price Forecasting
arXiv:2603.06726v1 Announce Type: new Abstract: Electricity market prices exhibit extreme volatility, nonlinearity, and non-stationarity, making accurate forecasting a significant challenge. While cutting-edge time series foundation models (TSFMs) effectively capture temporal dependencies, they typically underutilize cross-variate correlations and non-periodic patterns that...
Safe Transformer: An Explicit Safety Bit For Interpretable And Controllable Alignment
arXiv:2603.06727v1 Announce Type: new Abstract: Current safety alignment methods encode safe behavior implicitly within model parameters, creating a fundamental opacity: we cannot easily inspect why a model refuses a request, nor intervene when its safety judgments fail. We propose Safe...
Improved Constrained Generation by Bridging Pretrained Generative Models
arXiv:2603.06742v1 Announce Type: new Abstract: Constrained generative modeling is fundamental to applications such as robotic control and autonomous driving, where models must respect physical laws and safety-critical constraints. In real-world settings, these constraints rarely take the form of simple linear...
An Embodied Companion for Visual Storytelling
arXiv:2603.05511v1 Announce Type: cross Abstract: As artificial intelligence shifts from pure tool for delegation toward agentic collaboration, its use in the arts can shift beyond the exploration of machine autonomy toward synergistic co-creation. While our earlier robotic works utilized automation...
Spatiotemporal Heterogeneity of AI-Driven Traffic Flow Patterns and Land Use Interaction: A GeoAI-Based Analysis of Multimodal Urban Mobility
arXiv:2603.05581v1 Announce Type: cross Abstract: Urban traffic flow is governed by the complex, nonlinear interaction between land use configuration and spatiotemporally heterogeneous mobility demand. Conventional global regression and time-series models cannot simultaneously capture these multi-scale dynamics across multiple travel modes....
On the Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction
arXiv:2603.05532v1 Announce Type: cross Abstract: Despite continuing hype about the role of AI in drug discovery, no "AI-discovered drugs" have so far received regulatory approval. Here we assess one of the latest AI based tools in this domain. The ability...
Towards Efficient and Stable Ocean State Forecasting: A Continuous-Time Koopman Approach
arXiv:2603.05560v1 Announce Type: cross Abstract: We investigate the Continuous-Time Koopman Autoencoder (CT-KAE) as a lightweight surrogate model for long-horizon ocean state forecasting in a two-layer quasi-geostrophic (QG) system. By projecting nonlinear dynamics into a latent space governed by a linear...
Reasoning Models Struggle to Control their Chains of Thought
arXiv:2603.05706v1 Announce Type: new Abstract: Chain-of-thought (CoT) monitoring is a promising tool for detecting misbehaviors and understanding the motivations of modern reasoning models. However, if models can control what they verbalize in their CoT, it could undermine CoT monitorability. To...
The DSA's Blind Spot: Algorithmic Audit of Advertising and Minor Profiling on TikTok
arXiv:2603.05653v1 Announce Type: cross Abstract: Adolescents spend an increasing amount of their time in digital environments where their still-developing cognitive capacities leave them unable to recognize or resist commercial persuasion. Article 28(2) of the Digital Service Act (DSA) responds to...
MAPO: Mixed Advantage Policy Optimization for Long-Horizon Multi-Turn Dialogue
arXiv:2603.06194v1 Announce Type: new Abstract: Subjective multi-turn dialogue tasks, such as emotional support, require conversational policies that adapt to evolving user states and optimize long-horizon interaction quality. However, reinforcement learning (RL) for such settings remains challenging due to the absence...
Abductive Reasoning with Syllogistic Forms in Large Language Models
arXiv:2603.06428v1 Announce Type: new Abstract: Research in AI using Large-Language Models (LLMs) is rapidly evolving, and the comparison of their performance with human reasoning has become a key concern. Prior studies have indicated that LLMs and humans share similar biases,...
Sparse Crosscoders for diffing MoEs and Dense models
arXiv:2603.05805v1 Announce Type: new Abstract: Mixture of Experts (MoE) achieve parameter-efficient scaling through sparse expert routing, yet their internal representations remain poorly understood compared to dense models. We present a systematic comparison of MoE and dense model internals using crosscoders,...
MoE Lens -- An Expert Is All You Need
arXiv:2603.05806v1 Announce Type: new Abstract: Mixture of Experts (MoE) models enable parameter-efficient scaling through sparse expert activations, yet optimizing their inference and memory costs remains challenging due to limited understanding of their specialization behavior. We present a systematic analysis of...
DQE: A Semantic-Aware Evaluation Metric for Time Series Anomaly Detection
arXiv:2603.06131v1 Announce Type: new Abstract: Time series anomaly detection has achieved remarkable progress in recent years. However, evaluation practices have received comparatively less attention, despite their critical importance. Existing metrics exhibit several limitations: (1) bias toward point-level coverage, (2) insensitivity...
Fly in the Face of Bias: Algorithmic Bias in Law Enforcement’s Facial Recognition Technology and the Need for an Adaptive Legal Framework
THE FIGHT FOR PRIVACY: CALLING FOR BROAD ONLINE PRIVACY REFORM IN THE AGE OF BEING CHRONICALLY ONLINE - Minnesota Law Review
By Lea Chapoton, Volume 108 Staff Member In the wake of 2022’s Dobbs v. Jackson Women’s Health Organization[1] decision and the ensuing barrage of state laws limiting abortion access, online discussions surged with strategies for maintaining reproductive freedom in potentially...
Curbing Private Enforcement of the Voting Rights Act: Thoughts on Recent Developments
For decades, private plaintiffs have brought claims to enforce key provisions of the Voting Rights Act (VRA). Recent decisions have tossed out these claims on the ground that enforcement authority lies solely with the Attorney…The postCurbing Private Enforcement of the...
Banana republic: copyright law and the extractive logic of generative AI
Abstract This article uses Maurizio Cattelan’s Comedian, a banana duct-taped to a gallery wall, as a metaphor to examine the extractive dynamics of generative artificial intelligence (AI). It argues that the AI-driven creative economy replicates colonial patterns of appropriation, transforming...
D-BIAS: A Causality-Based Human-in-the-Loop System for Tackling Algorithmic Bias
With the rise of AI, algorithms have become better at learning underlying patterns from the training data including ingrained social biases based on gender, race, etc. Deployment of such algorithms to domains such as hiring, healthcare, law enforcement, etc. has...
Anti-Domination and Administration
The foundations of the administrative state are being reshaped, both by the continuing transformations of administrative law doctrine by the courts and by the ambitions for restructuring the executive branch among the current presidential administration. But at the same time,...
Fourth Amendment Equilibrium Adjustment in an Age of Technological Upheaval
The Digital Fourth Amendment is written by Professor Orin Kerr, one of the country’s foremost authorities on the Fourth Amendment, electronic privacy, and criminal procedure. Kerr’s work has been deeply influential in shaping how courts are looking at and deciding...
Geometric Conservation Law and Its Application to Flow Computations on Moving Grids
Boundary-conforming coordinate transformations are used widely to map a flow region onto a computational space in which a finite-difference solution to the differential flow conservation laws is carried out. This method entails difficulties with maintenance of global conservation and with...
Criticality, the Area Law, and the Computational Power of Projected Entangled Pair States
The projected entangled pair state (PEPS) representation of quantum states on two-dimensional lattices induces an entanglement based hierarchy in state space. We show that the lowest levels of this hierarchy exhibit a very rich structure including states with critical and...
Data augmentation for fairness-aware machine learning
Researchers and practitioners in the fairness community have highlighted the ethical and legal challenges of using biased datasets in data-driven systems, with algorithmic bias being a major concern. Despite the rapidly growing body of literature on fairness in algorithmic decision-making,...
Russian Court Decisions Data Analysis Using Distributed Computing and Machine Learning to Improve Lawmaking and Law Enforcement
This article describes the study results of semi-structured data processing and analysis of the Russian court decisions (almost 30 million) using distributed cluster-computing framework and machine learning. Spark was used for data processing and decisions trees were used for analysis....