Large Language Models

I usually cycle through my years-old playlists, but I tried AI-generated ones for a weekend. I found being specific is key.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. The gig workers who ar

Considering ditching ChatGPT for Claude? I tested both on the same 10 tasks. Here's which came out on top.

The Copenhagen-based health AI company built Symphony on peer-reviewed research from the largest medical coding study of its kind, treating coding as a reasonin

Want to know what our reviewers have actually tested and picked as the best TVs, headphones, and laptops? Ask ChatGPT, and it'll give you the wrong answers.

Earlier this year, Mitchell Hashimoto coined the term "harness engineering" — the discipline of building everything around the model that makes an AI agent actu

I Built an AI PPT Maker and Resume Builder Website built a website that helps students and professionals create PowerPoint presentations and resumes using AI in

Listicles get cited by AI engines 21.9% of the time. Articles follow at 16.7%. Product pages hit 13.7%. This is the first hard data on what content formats Chat

Every AI agent can process information. Most can reason through complex problems. But theres a fundamental gap thats becoming impossible to ignore: agents cant

AI uses text to converse on mental health aspects. We are moving to multimodal interactions. Fusion is crucial. Especially for mental health chats. An AI Inside

Apple introduces two foundation language models behind Apple Intelligence: a ~3B on-device model optimized for Apple silicon and a scalable server model using a

Anthropic’s Mythos leak exposed alarming security lapses, raising trust concerns as powerful AI models emerge.

arXiv:2603.28902v1 Announce Type: new Abstract: Charts are central to analytical reasoning, yet existing benchmarks for chart understanding focus almost exclusi

arXiv:2603.28955v1 Announce Type: new Abstract: This paper presents the World-Action Model (WAM), an action-regularized world model that jointly reasons over fu

arXiv:2603.28990v1 Announce Type: new Abstract: How much autonomy can multi-agent LLM systems sustain -- and what enables it? We present a 25,000-task computati

arXiv:2603.29075v1 Announce Type: new Abstract: The way we're thinking about generative AI right now is fundamentally individual. We see this not just in how us

arXiv:2603.29085v1 Announce Type: new Abstract: Large language models (LLMs) remain brittle on multi-hop question answering (MHQA), where answering requires com

arXiv:2603.29112v1 Announce Type: new Abstract: We introduce GISTBench, a benchmark for evaluating Large Language Models' (LLMs) ability to understand users fro

arXiv:2603.29142v1 Announce Type: new Abstract: Formative feedback is central to effective learning, yet providing timely, individualised feedback at scale rema

arXiv:2603.29149v1 Announce Type: new Abstract: Access to the most up-to-date information on medical countermeasures is important for the research and developme

arXiv:2603.29161v1 Announce Type: new Abstract: Modern web scraping struggles with dynamic, interactive websites that require more than static HTML parsing. Cur

arXiv:2603.29206v1 Announce Type: new Abstract: Routing is widely used to scale large language models, from Mixture-of-Experts gating to multi-model/tool select

arXiv:2603.29211v1 Announce Type: new Abstract: In recent years, multimodal large models have continued to improve on general benchmarks. However, in real-world

arXiv:2603.29231v1 Announce Type: new Abstract: Existing benchmarks measure capability -- whether a model succeeds on a single attempt -- but production deploym

arXiv:2603.29262v1 Announce Type: new Abstract: Grokking in modular arithmetic has established itself as the quintessential fruit fly experiment, serving as a c

arXiv:2603.29366v1 Announce Type: new Abstract: Prior authorization remains one of the most burdensome administrative processes in U.S. healthcare, consuming bi

arXiv:2603.29491v1 Announce Type: new Abstract: In the evaluation of attribution quality, the quantitative assessment of explanation legibility is particularly

arXiv:2603.29500v1 Announce Type: new Abstract: Large language models (LLMs) have recently demonstrated impressive performance on complex, multi-step reasoning

arXiv:2603.29557v1 Announce Type: new Abstract: Scientific idea generation (SIG) is critical to AI-driven autonomous research, yet existing approaches are often

arXiv:2603.29681v1 Announce Type: new Abstract: The common claim that generative AI simply amplifies the Dunning-Kruger effect is too coarse to capture the avai

arXiv:2603.29735v1 Announce Type: new Abstract: The evolution of intelligence in artificial systems provides a unique opportunity to identify universal computat

arXiv:2603.29791v1 Announce Type: new Abstract: Although many AI applications of interest require specialized multi-modal models, relevant data to train such mo

arXiv:2603.29848v1 Announce Type: new Abstract: We introduce a comprehensive validation framework for LLM-based agentic systems that provides systematic diagnos

arXiv:2603.29871v1 Announce Type: new Abstract: In user-agent interaction scenarios such as recommendation, brainstorming, and code suggestion, Large Language M

arXiv:2603.29902v1 Announce Type: new Abstract: Interleaved text-and-image generation represents a significant frontier for Multimodal Large Language Models (ML

arXiv:2603.29908v1 Announce Type: new Abstract: Trajectory planning for autonomous driving increasingly leverages large language models (LLMs) for commonsense r

arXiv:2603.29915v1 Announce Type: new Abstract: Post-hoc explanation methods are widely used to interpret black-box predictions, but their generation is often c

arXiv:2603.29953v1 Announce Type: new Abstract: How reliably can structured intent representations preserve user goals across different AI models, languages, an

arXiv:2603.30031v1 Announce Type: new Abstract: Current autonomous AI agents, driven primarily by Large Language Models (LLMs), operate in a state of cognitive

arXiv:2603.27006v1 Announce Type: cross Abstract: Large language models produce em dashes at varying rates, and the observation that some models "overuse" them

arXiv:2603.28795v1 Announce Type: cross Abstract: We address LLM serving workloads where repeated requests share a common solution structure but differ in local

arXiv:2603.28817v1 Announce Type: cross Abstract: Small Language Models (SLMs) are emerging as efficient and economically viable alternatives to Large Language

arXiv:2603.28845v1 Announce Type: cross Abstract: Deploying foundation models is increasingly constrained by memory footprint, latency, and hardware costs. Post

arXiv:2603.28858v1 Announce Type: cross Abstract: Continual pre-training is widely used to adapt LLMs to target languages and domains, yet the mixture ratio of

arXiv:2603.28921v1 Announce Type: cross Abstract: Standard neural network training uses constant momentum (typically 0.9), a convention dating to 1964 with limi

arXiv:2603.28925v1 Announce Type: cross Abstract: Safety fine-tuning in Large Language Models (LLMs) seeks to suppress potentially harmful forms of mind-attribu

arXiv:2603.28959v1 Announce Type: cross Abstract: The exploration-exploitation trade-off is central to sequential decision-making and black-box optimization, ye