Alignment, bias, regulation, and responsible AI
![[2511.21331] The More, the Merrier: Contrastive Fusion for Higher-Order Multimodal Alignment](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
Abstract page for arXiv paper 2511.21331: The More, the Merrier: Contrastive Fusion for Higher-Order Multimodal Alignment
Abstract page for arXiv paper 2509.22367: What Is The Political Content in LLMs' Pre- and Post-Training Data?
Abstract page for arXiv paper 2507.22264: SmartCLIP: Modular Vision-language Alignment with Identification Guarantees
The article presents a Decision Quality Evaluation Framework developed at Pinterest to enhance content moderation by evaluating the quali...
The paper introduces mini-vec2vec, an efficient method for aligning text embedding spaces using linear transformations, significantly imp...
This paper explores vulnerabilities in diffusion language models (DLMs) related to priming attacks and proposes a novel safety alignment ...
The paper presents FlowDrive, a trajectory planning method that addresses data imbalance in driving datasets by using moderated flow matc...
The paper presents FineMuSe, a new dataset for detecting nuanced sexism in social media videos, addressing the limitations of binary clas...
The paper presents VLCE, a framework that enhances image description for disaster assessment by integrating external semantic knowledge, ...
This paper explores enhancing user interface (UI) generation models by incorporating designer feedback, demonstrating improved performanc...
The article examines the impact of strategic AI disclosure in crowdfunding, revealing that mandatory disclosure can significantly reduce ...
This paper presents a content-based framework for cybersecurity refusal decisions in large language models, emphasizing the need for expl...
This article presents a data-driven framework for estimating human muscular fatigue during collaborative robotic tasks using machine lear...
The paper presents a differentially private sampling algorithm, Reveal-or-Obscure (ROO), for generating samples from discrete distributio...
This paper discusses the security vulnerabilities of self-evolving LLM agents, introducing the concept of 'Zombie Agents' that can be cov...
The paper presents STAPO, a novel approach to stabilize reinforcement learning in large language models by silencing rare spurious tokens...
The paper introduces VLM-DEWM, a novel cognitive architecture designed to enhance vision-language planning in manufacturing by addressing...
This paper introduces the Minimum Path Variance (MinPV) Principle, addressing the paradox of score-based methods in machine learning by m...
Green-NAS presents a multi-objective neural architecture search framework aimed at optimizing weather forecasting models for low-resource...
SecCodeBench-V2 is a benchmark for evaluating LLMs' ability to generate secure code, featuring 98 scenarios across five programming langu...
This article examines various algorithmic approaches to opinion selection in online deliberation, highlighting the trade-offs between div...
The paper introduces ARGUS, a novel framework for detecting distributional drift in high-dimensional data streams, emphasizing geometric ...
This paper presents a systematic evaluation of learning-based similarity techniques for malware detection, comparing various methods unde...
Get the latest news, tools, and insights delivered to your inbox.
Daily or weekly digest • Unsubscribe anytime