洞察 AI 演进与全球贸易脉动，以全息视野重塑跨域情报流。
Decoding AI evolution and global trade rhythms. High-fidelity intelligence, redefined.
Formal Mechanisms for Market Stability in Self-Interested Agent Societies: A Marketplace Simulation Study
自私的代理人,不受约束,倾向于在重复的社会困境中叛逃,，导致贸易合作收益崩溃。这 ...
Self-interested agents, left unconstrained, tend toward defection in repeated social dilemmas, causing cooperative gains from trade to collapse. This ...
SolarChain-Eval: A Physics-Constrained Benchmark for Trustworthy Economic Agents in Decentralized Energy Markets
随着代理人工智能系统越来越多地应用于网络物理环境,，他们的评估需要评估任务绩效和信任......
As agentic AI systems are increasingly applied to cyber-physical environments, their evaluation requires assessment of both task performance and trust...
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
在长期任务中，, 决策相关状态通常分散在不断扩展的轨迹, 上，而行动代理必须将其浮现出来并采取行动。作为...
In long-horizon tasks, decision-relevant state is often scattered across an expanding trajectory, while the action agent must surface it and act. As t...
The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs
训练后量化已广泛用于压缩大型语言模型，使其可部署在资源受限的设备上。然而,...
Post-Training Quantization has become widely used to compress large language models to make them deployable on resource-constrained devices. However, ...
Workflow as Knowledge: Semantic Persistence for LLM-Mediated Workflows
大型语言模型 (LLM) 应用程序越来越多地使用显式工作流程来进行工具使用, 检索, 分支, 检查点, 和人工审批。埃西...
Large language model (LLM) applications increasingly use explicit workflows for tool use, retrieval, branching, checkpointing, and human approval. Exi...
AUTOPILOT VQA: Benchmarking Vision-Language Models for Incident-Centric Dashcam Understanding
视觉语言模型,大型语言模型,和多模态大型语言模型的最新进展改善了自动驾驶任务，例如......
Recent advances in Vision-Language Models, Large Language Models, and Multimodal Large Language Models have improved autonomous driving tasks such as ...
Using AI-based Learning Assistants in Higher Education: A Large-Scale Descriptive Analysis
在这项研究,中，我们对基于人工智能的学习助手(Syntea)在高等教育中的使用进行了大规模的描述性分析。基于对象...
In this study, we present a large-scale descriptive analysis of the use of an AI-based learning assistant (Syntea) in higher education. Based on objec...
Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation
科学思想很少是从白纸开始的。他们继承了机制, 修复已知的局限性, 并重新组合早期工作, 的各个部分，就像双...
Scientific ideas rarely start from a blank page. They inherit mechanisms, repair known limitations, and recombine pieces of earlier work, much like bi...
Institutional Red-Teaming: Deployment Rules, Not Just Models, Causally Shape Multi-Agent AI Safety
我们引入机构红队,一种用于测试多智能体AI中的部署规则的评估方法:持有代理,目标,和任务...
We introduce institutional red-teaming, an evaluation methodology for testing deployment rules in multi-agent AI: hold the agents, objectives, and tas...
Tiny robot boats build floating structures
大多数人认为海滨是城市的边缘。麻省理工学院的一组研究人员将其视为一个充满活力的, 乐高式建筑工地。他们的新系统...
Most people think of the waterfront as the edge of the city. A team of MIT researchers sees it as a dynamic, Lego-like construction site. Their new sy...