VLMShield: Efficient and Robust Defense of Vision-Language Models against Malicious Prompts
arXiv:2604.06502v1 Announce Type: new Abstract: Vision-Language Models (VLMs) face significant safety vulnerabilities from malicious prompt attacks due to weakened alignment during visual integration. Existing defenses suffer from efficiency and robustness. To address these challenges, we first propose the Multimodal Aggregated...
Beyond Facts: Benchmarking Distributional Reading Comprehension in Large Language Models
arXiv:2604.06201v1 Announce Type: new Abstract: While most reading comprehension benchmarks for LLMs focus on factual information that can be answered by localizing specific textual evidence, many real-world tasks require understanding distributional information, such as population-level trends and preferences expressed across...
Stop Fixating on Prompts: Reasoning Hijacking and Constraint Tightening for Red-Teaming LLM Agents
arXiv:2604.05549v1 Announce Type: new Abstract: With the widespread application of LLM-based agents across various domains, their complexity has introduced new security threats. Existing red-team methods mostly rely on modifying user prompts, which lack adaptability to new data and may impact...
AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discovery
arXiv:2604.05550v1 Announce Type: new Abstract: Artificial intelligence research increasingly depends on prolonged cycles of reproduction, debugging, and iterative refinement to achieve State-Of-The-Art (SOTA) performance, creating a growing need for systems that can accelerate the full pipeline of empirical model optimization....
Thinking Diffusion: Penalize and Guide Visual-Grounded Reasoning in Diffusion Multimodal Language Models
arXiv:2604.05497v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) are emerging as promising alternatives to autoregressive (AR) LLMs. Recently, this paradigm has been extended to multimodal tasks, leading to the development of diffusion multimodal large language models (dMLLMs). These...
LMI-Net: Linear Matrix Inequality--Constrained Neural Networks via Differentiable Projection Layers
arXiv:2604.05374v1 Announce Type: new Abstract: Linear matrix inequalities (LMIs) have played a central role in certifying stability, robustness, and forward invariance of dynamical systems. Despite rapid development in learning-based methods for control design and certificate synthesis, existing approaches often fail...
TDA-RC: Task-Driven Alignment for Knowledge-Based Reasoning Chains in Large Language Models
arXiv:2604.04942v1 Announce Type: new Abstract: Enhancing the reasoning capability of large language models (LLMs) remains a core challenge in natural language processing. The Chain-of-Thought (CoT) paradigm dominates practical applications for its single-round efficiency, yet its reasoning chains often exhibit logical...
TRACE: Capability-Targeted Agentic Training
arXiv:2604.05336v1 Announce Type: new Abstract: Large Language Models (LLMs) deployed in agentic environments must exercise multiple capabilities across different task instances, where a capability is performing one or more actions in a trajectory that are necessary for successfully solving a...
DIA-HARM: Dialectal Disparities in Harmful Content Detection Across 50 English Dialects
arXiv:2604.05318v1 Announce Type: new Abstract: Harmful content detectors-particularly disinformation classifiers-are predominantly developed and evaluated on Standard American English (SAE), leaving their robustness to dialectal variation unexplored. We present DIA-HARM, the first benchmark for evaluating disinformation detection robustness across 50 English...
Faster Superword Tokenization
arXiv:2604.05192v1 Announce Type: new Abstract: Byte Pair Encoding (BPE) is a widely used tokenization algorithm, whose tokens cannot extend across pre-tokenization boundaries, functionally limiting it to representing at most full words. The BoundlessBPE and SuperBPE algorithms extend and improve BPE...
ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspaces
arXiv:2604.05172v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly deployed to automate productivity tasks (e.g., email, scheduling, document management), but evaluating them on live services is risky due to potentially irreversible changes. Existing benchmarks rely on simplified...
Vehicle-as-Prompt: A Unified Deep Reinforcement Learning Framework for Heterogeneous Fleet Vehicle Routing Problem
arXiv:2604.05195v1 Announce Type: new Abstract: Unlike traditional homogeneous routing problems, the Heterogeneous Fleet Vehicle Routing Problem (HFVRP) involves heterogeneous fixed costs, variable travel costs, and capacity constraints, rendering solution quality highly sensitive to vehicle selection. Furthermore, real-world logistics applications often...
Expectation Maximization (EM) Converges for General Agnostic Mixtures
arXiv:2604.05842v1 Announce Type: new Abstract: Mixture of linear regression is well studied in statistics and machine learning, where the data points are generated probabilistically using $k$ linear models. Algorithms like Expectation Maximization (EM) may be used to recover the ground...
Energy-Based Dynamical Models for Neurocomputation, Learning, and Optimization
arXiv:2604.05042v1 Announce Type: new Abstract: Recent advances at the intersection of control theory, neuroscience, and machine learning have revealed novel mechanisms by which dynamical systems perform computation. These advances encompass a wide range of conceptual, mathematical, and computational ideas, with...
EvolveRouter: Co-Evolving Routing and Prompt for Multi-Agent Question Answering
arXiv:2604.05149v1 Announce Type: new Abstract: Large language model agents often exhibit complementary strengths, making routing a promising approach for multi-agent question answering. However, existing routing methods remain limited in two important ways: they typically optimize over a fixed pool of...
The 14th Amendment’s citizenship clause is not trapped in amber: a reflection on oral argument
While I have written multiple posts for SCOTUSblog on birthright citizenship, a substantial part of my practice is litigating Second Amendment claims. In light of that experience, I was struck […]The postThe 14th Amendment’s citizenship clause is not trapped in...
What oral arguments and opinion authorships can actually tell us
Empirical SCOTUS is a recurring series by Adam Feldman that looks at Supreme Court data, primarily in the form of opinions and oral arguments, to provide insights into the justices’ decision making and […]The postWhat oral arguments and opinion authorships...
Intel signs on to Elon Musk’s Terafab chips project
Intel will join SpaceX and Tesla in an effort to build a new U.S. semiconductor factory in Texas, although the scope of its contributions are unclear.
The Higher Education Accommodation Mistake
Made in the U.S.A.: The Constitutional Crisis Behind America’s Arms Export Regime
Rethinking the Key Role of Private Antitrust Enforcement
Episode 42: Russia, Imperial Continuities and Histories of International Law - EJIL: The Podcast!
From Model-Based Screening to Data-Driven Surrogates: A Multi-Stage Workflow for Exploring Stochastic Agent-Based Models
arXiv:2604.03350v1 Announce Type: new Abstract: Systematic exploration of Agent-Based Models (ABMs) is challenged by the curse of dimensionality and their inherent stochasticity. We present a multi-stage pipeline integrating the systematic design of experiments with machine learning surrogates. Using a predator-prey...
Many Preferences, Few Policies: Towards Scalable Language Model Personalization
arXiv:2604.04144v1 Announce Type: new Abstract: The holy grail of LLM personalization is a single LLM for each user, perfectly aligned with that user's preferences. However, maintaining a separate LLM per user is impractical due to constraints on compute, memory, and...
Evaluating Artificial Intelligence Through a Christian Understanding of Human Flourishing
arXiv:2604.03356v1 Announce Type: new Abstract: Artificial intelligence (AI) alignment is fundamentally a formation problem, not only a safety problem. As Large Language Models (LLMs) increasingly mediate moral deliberation and spiritual inquiry, they do more than provide information; they function as...
Supervised Dimensionality Reduction Revisited: Why LDA on Frozen CNN Features Deserves a Second Look
arXiv:2604.03928v1 Announce Type: new Abstract: Effective ride-hailing dispatch requires anticipating demand patterns that vary substantially across time-of-day, day-of-week, season, and special events. We propose a regime-calibrated approach that (i) segments historical trip data into demand regimes, (ii) matches the current...
Position: Science of AI Evaluation Requires Item-level Benchmark Data
arXiv:2604.03244v1 Announce Type: new Abstract: AI evaluations have become the primary evidence for deploying generative AI systems across high-stakes domains. However, current evaluation paradigms often exhibit systemic validity failures. These issues, ranging from unjustified design choices to misaligned metrics, remain...
Earth Embeddings Reveal Diverse Urban Signals from Space
arXiv:2604.03456v1 Announce Type: new Abstract: Conventional urban indicators derived from censuses, surveys, and administrative records are often costly, spatially inconsistent, and slow to update. Recent geospatial foundation models enable Earth embeddings, compact satellite image representations transferable across downstream tasks, but...
Understanding When Poisson Log-Normal Models Outperform Penalized Poisson Regression for Microbiome Count Data
arXiv:2604.03853v1 Announce Type: new Abstract: Multivariate count models are often justified by their ability to capture latent dependence, but researchers receive little guidance on when this added structure improves on simpler penalized marginal Poisson regression. We study this question using...
GeoBrowse: A Geolocation Benchmark for Agentic Tool Use with Expert-Annotated Reasoning Traces
arXiv:2604.04017v1 Announce Type: new Abstract: Deep research agents integrate fragmented evidence through multi-step tool use. BrowseComp offers a text-only testbed for such agents, but existing multimodal benchmarks rarely require both weak visual cues composition and BrowseComp-style multi-hop verification. Geolocation is...