891020 results (page 23 of 35641)
-
MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-Judge
Multimodal Large Language Models (MLLMs) have been increasingly used as automatic evaluators-a paradigm known as MLLM-as-a-Judge. However, their reliability and vulnerabilities to biases remain underexplored. We find that many MLLM judges fail to reliably integrate key visual or textual cues, yielding unreliable evaluations when evidence is missing or mismatched, and exhibiting instability under s…
-
Audit-or-Cast: Enforcing Honest Elections with Privacy-Preserving Public Verification
Electronic voting systems must balance public verifiability with voter privacy and coercion resistance. Existing cryptographic protocols typically achieve end-to-end verifiability by revealing vote distributions, relying on trusted clients, or enabling transferable receipts - design choices that often compromise trust or privacy in real-world deployments. We present ACE, a voting protocol that r…
-
Does "Do Differentiable Simulators Give Better Policy Gradients?'' Give Better Policy Gradients?
In policy gradient reinforcement learning, access to a differentiable model enables 1st-order gradient estimation that accelerates learning compared to relying solely on derivative-free 0th-order estimators. However, discontinuous dynamics cause bias and undermine the effectiveness of 1st-order estimators. Prior work addressed this bias by constructing a confidence interval around the REINFORCE 0t…
-
FreezeEmpath: Efficient Training for Empathetic Spoken Chatbots with Frozen LLMs
Empathy is essential for fostering natural interactions in spoken dialogue systems, as it enables machines to recognize the emotional tone of human speech and deliver empathetic responses. Recent research has made significant progress in developing empathetic spoken chatbots based on large language models (LLMs). However, several challenges still exist when training such models, including reliance…
-
State Transfer Reveals Reuse in Controlled Routing
Prompt-based interventions can change model behavior, but trained success alone does not identify where the behaviorally relevant state is represented. We study this question in controlled routing tasks using interfaces chosen on support data, held-out query evaluation, and matched necessity, sufficiency, and wrong-interface controls. On GPT-2 triop, an early interface supports exact transfer unde…
-
Geometry-Aware Networking for Low-Altitude Economy: Movable Antennas in Space-Air-Ground Integrated Systems
Space--air--ground integrated networks (SAGINs) are emerging as a key foundation for future non-terrestrial networks (NTNs) and low-altitude economy services. However, their performance is increasingly limited not only by communication resources, but by the inability to adapt to rapidly changing spatial geometry. Here, spatial geometry refers to the relative configuration among network nodes, obst…
-
Leveraging AI for Direct Bystander Intervention Against Cyberbullying
Cyberbullying is a pervasive problem in online environments, causing substantial psychological harm to victims. Although bystander intervention has proven effective in mitigating its impact, motivating bystanders to engage in direct intervention remains a persistent challenge. Studies have suggested that difficulties in intervention skills and defending self-efficacy hinder bystanders from initiat…
-
mlr3torch: A Deep Learning Framework in R based on mlr3 and torch
Deep learning (DL) has become a cornerstone of modern machine learning (ML) praxis. We introduce the R package mlr3torch, which is an extensible DL framework for the mlr3 ecosystem. It is built upon the torch package, and simplifies the definition, training, and evaluation of neural networks for both tabular data and generic tensors (e.g., images) for classification and regression. The package imp…
-
AI-based Waste Mapping for Addressing Climate-Exacerbated Flood Risk
Urban flooding is a growing climate change-related hazard in rapidly expanding African cities, where inadequate waste management often blocks drainage systems and amplifies flood risks. This study introduces an AI-powered urban waste mapping workflow that leverages openly available aerial and street-view imagery to detect municipal solid waste at high resolution. Applied in Dar es Salaam, Tanzania…
-
On the relative CNO underabundance in quasar absorption systems at $z \sim 3$ arising from Population III enrichment and attenuation by intermediate-mass black holes and primordial baryon accretion
This article uses an adapted version of the semi-analytical model of cosmic chemical enrichment developed by \citet{Corazza_2022} to reproduce the observed abundances of C, N, and O in absorption systems of quasar spectra (ASQS) at $z \gtrsim 3-6$, addressing an overproduction issue of the abovementioned elements. We address this discrepancy by updating the cosmic star formation rate (CSFR) and …
-
Informativity of Data-Knowledge Pairs for Lyapunov Equations
In the past few years, data informativity with prior knowledge has attracted increasing attention. This line of research aims to characterize a dataset on a dynamical system that enables system analysis or design only by the dataset and given prior knowledge on the system. In this paper, we investigate such a characterization for the data-driven problem of computing a unique solution to Lyapunov e…
-
Attention-ResUNet for Automated Fetal Head Segmentation
Automated fetal head segmentation in ultrasound images is critical for accurate biometric measurements in prenatal care. While existing deep learning approaches have achieved a reasonable performance, they struggle with issues like low contrast, noise, and complex anatomical boundaries which are inherent to ultrasound imaging. This paper presents Attention-ResUNet. It is a novel architecture that …
-
The Magnitude of Dominated Sets: A Pareto Compliant Indicator Grounded in Metric Geometry
We investigate \emph{magnitude} as a new unary and strictly Pareto-compliant quality indicator for finite approximation sets to the Pareto front in multiobjective optimization. Magnitude originates in enriched category theory and metric geometry, where it is a notion of size or point content for compact metric spaces and a generalization of cardinality. For dominated regions in the \(\ell_1\) box …
-
Modular Representation Compression: Adapting LLMs for Efficient and Effective Recommendations
Recently, large language models (LLMs) have advanced recommendation systems (RSs), and recent works have begun to explore how to integrate LLMs into industrial RSs. While most approaches deploy LLMs offline to generate and pre-cache augmented representations for RSs, high-dimensional representations from LLMs introduce substantial storage and computational costs. Thus, it is crucial to compress LL…
-
Region-Grounded Report Generation for 3D Medical Imaging: A Fine-Grained Dataset and Graph-Enhanced Framework
Automated medical report generation for 3D PET/CT imaging is fundamentally challenged by the high-dimensional nature of volumetric data and a critical scarcity of annotated datasets, particularly for low-resource languages. Current black-box methods map whole volumes to reports, ignoring the clinical workflow of analyzing localized Regions of Interest (RoIs) to derive diagnostic conclusions. In th…
-
Distributional Off-Policy Evaluation with Deep Quantile Process Regression
This paper investigates the off-policy evaluation (OPE) problem from a distributional perspective. Rather than focusing solely on the expectation of the total return, as in most existing OPE methods, we aim to estimate the entire return distribution. To this end, we introduce a quantile-based approach for OPE using deep quantile process regression, presenting a novel algorithm called Deep Quantile…
-
Frugal Geofencing via Energy-aware Sensing and Reporting
Timely and accurate monitoring in geofencing scenarios is challenging when relying on ultra-low power Internet of Things devices (IoTDs) powered by energy harvesting (EH). This is mainly because frequent wake-ups for data acquisition and data uploading may quickly deplete their limited energy buffer. Conventional grid-like IoT deployments overlook these limitations and merely rely on continuously …
-
Leader-Follower Formation Control Using Differential Drag and Effective Surface Regulation
The growing interest in space activities has led to the emergence of new space operators and innovative mission concepts. Small satellites such as CubeSats reduce mission costs and are typically deployed in constellations or formation flights. Since they are often propulsionless, passive orbital control strategies are the standard, primarily through differential drag achieved via attitude control …
-
Semi-Blind Receivers for RIS-Aided Fluid Antenna Systems
Reconfigurable intelligent surfaces (RISs) and fluid antennas (FAs) are key technologies for enhancing spatial degrees of freedom in future wireless networks. However, channel acquisition in RIS-aided FA systems is challenging as cascaded links depend on time-varying antenna-port selections and RIS configurations, leading to high training overhead in conventional pilot-based methods. We propose a …
-
AQPIM: Breaking the PIM Capacity Wall for LLMs with In-Memory Activation Quantization
Processing-in-Memory (PIM) architectures offer a promising solution to the memory bottlenecks in data-intensive machine learning, yet often overlook the growing challenge of activation memory footprint. Conventional PIM approaches struggle with massive KV cache sizes generated in long-context scenarios by Transformer-based models, frequently exceeding PIM's limited memory capacity, while technique…
-
Equation of State for warm Neutron Star outer crusts
We describe the equation of state (EoS) of a warm ion plasma as obtained by performing microscopic many-body simulations using Molecular Dynamics computational techniques. Using the cold one-component plasma (OCP) composition in the Neutron Star (NS) outer crust assumed in Murarka et al. (2022) with a representative heavy nucleus for each density, we refine previous calculations. We include electr…
-
Soft Label Pruning and Quantization for Large-Scale Dataset Distillation
Large-scale dataset distillation requires storing auxiliary soft labels that can be 30-40x larger on ImageNet-1K and 200x larger on ImageNet-21K than the condensed images, undermining the goal of dataset compression. We identify two fundamental issues necessitating such extensive labels: (1) insufficient image diversity, where high within-class similarity in synthetic images requires extensive aug…
-
Can LLM-Generated Text Empower Surgical Vision-Language Pre-training?
Recent advancements in self-supervised learning have led to powerful surgical vision encoders capable of spatiotemporal understanding. However, extending these visual foundations to multi-modal reasoning tasks is severely bottlenecked by the prohibitive cost of expert textual annotations. To overcome this scalability limitation, we introduce \textbf{LIME}, a large-scale multi-modal dataset derived…
-
Multi-Agent Systems: From Classical Paradigms to Large Foundation Model-Enabled Futures
With the rapid advancement of artificial intelligence, multi-agent systems (MASs) are evolving from classical paradigms toward architectures built upon large foundation models (LFMs). This survey provides a systematic review and comparative analysis of classical MASs (CMASs) and LFM-based MASs (LMASs). First, within a closed-loop coordination framework, CMASs are reviewed across four fundamental d…
-
Do time delay effects explain galactic velocity profiles?
Using the gravitoelectromagnetic analogy for weak gravitational fields, we critique explanations of galactic velocity profiles that invoke time delay effects (i.e. "retarded gravity"). For isotropic, time-dependent matter currents, we show within this framework that the force exerted on an orbiting body is Newtonian and due only to the instantaneous ambient matter configuration -- there are no tim…