📰 Reads

arXiv:2604.24881v1 Announce Type: new Abstract: Multi-agent debate has been shown to improve reasoning in large language models (LLMs). However, it is compute-i

arXiv:2604.24933v1 Announce Type: new Abstract: General audio foundation models have recently achieved remarkable progress, enabling strong performance across d

arXiv:2604.24983v1 Announce Type: new Abstract: Existing white-box jailbreak attacks against aligned LLMs typically append discrete adversarial suffixes to the

arXiv:2604.24987v1 Announce Type: new Abstract: Chart-to-table translation converts chart images into structured tabular data. Accurate translation is crucial f

arXiv:2604.24996v1 Announce Type: new Abstract: As Large Language Models (LLMs) advance, personalization has become a key mechanism for tailoring outputs to ind

arXiv:2604.25000v1 Announce Type: new Abstract: Recent work has framed intelligence in verifiable tasks as reducing time-to-solution through learned structure a

arXiv:2604.25040v1 Announce Type: new Abstract: We propose a per-task leverage ratio for human-agent collaboration: human work displaced by an agent, divided by

arXiv:2604.25077v1 Announce Type: new Abstract: Weak-to-strong alignment offers a promising route to scalable supervision, but it can fail when a strong model b

arXiv:2604.25083v1 Announce Type: new Abstract: Rapid advances in Large Language Models (LLMs) create new opportunities by enabling efficient exploration of bro

arXiv:2604.25088v1 Announce Type: new Abstract: Language Model (LM)-based agents remain largely untested in mixed-motive settings where agents must leverage sho

arXiv:2604.25098v1 Announce Type: new Abstract: While current Large Language Models (LLMs) exhibit remarkable reasoning capabilities through test-time compute s

arXiv:2604.25149v1 Announce Type: new Abstract: LLMs deployed for natural-language querying of analytical databases suffer from two intertwined failures - incor

arXiv:2604.25166v1 Announce Type: new Abstract: We demonstrate that a small transformer can learn to execute programs in MicroPy, a simplified yet computational

arXiv:2604.25167v1 Announce Type: new Abstract: While mechanistic interpretability tools like Sparse Autoencoders (SAEs) can uncover meaningful features within

arXiv:2604.25220v1 Announce Type: new Abstract: Data videos are a powerful medium for visual data based storytelling, combining animated, chart-centric visualiz

arXiv:2604.25224v1 Announce Type: new Abstract: Long-horizon investment decisions create a pre-realization evaluation problem: realized returns are the eventual

arXiv:2604.25256v1 Announce Type: new Abstract: Autonomous scientific research is significantly advanced thanks to the development of AI agents. One key step in

arXiv:2604.25345v1 Announce Type: new Abstract: Agentic AI systems are increasingly being integrated into scientific workflows, yet their behavior under realist

arXiv:2604.25369v1 Announce Type: new Abstract: Over the past few decades, machine learning has been widely used to learn complex tasks. Reinforcement Learning

arXiv:2604.25419v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) enhances the reasoning of large language models (LLMs), bu

arXiv:2604.25435v1 Announce Type: new Abstract: Source-free test-time adaptation (TTA) is appealing for mobile and wearable sensing because it enables on-device

arXiv:2604.25472v1 Announce Type: new Abstract: The need to evaluate instructional materials for K-12 science education has become increasingly important, as mo

arXiv:2604.25496v1 Announce Type: new Abstract: Offline zero-shot reinforcement learning (RL) aims to learn agents that optimize unseen reward functions without

arXiv:2604.25512v1 Announce Type: new Abstract: Phishing detection systems are predominantly rely on statistical machine learning models, which often lack conte