Curating the frontier of Artificial Intelligence

DeepWeb-Bench: A Deep Research Benchmark Demanding Massive Cross-Source Evidence and Long-Horizon Derivation

深入研究,，其中代理搜索开放网络, 收集证据, 并通过扩展推理得出答案, 是一个突出的用例...

Deep research, in which an agent searches the open web, collects evidence, and derives an answer through extended reasoning, is a prominent use case f...

Technology usually creates jobs for young, skilled workers. Will AI do the same?

在任何给定时间, 技术对就业有两个作用: 它取代了传统工作, 并且创造了新的工作领域。机器取代农民,...

At any given time, technology does two things to employment: It replaces traditional jobs, and it creates new lines of work. Machines replace farmers,...

A Methodology for Selecting and Composing Runtime Architecture Patterns for Production LLM Agents

生产 LLM 代理将随机模型输出与确定性软件系统, 相结合，但两者之间的边界很少被视为一个界限...

Production LLM agents combine stochastic model outputs with deterministic software systems, yet the boundary between the two is rarely treated as a fi...

Building AI models that understand chemical principles

在所有可能的化合物中, it的估计1020到1060之间可能具有作为小分子药物的潜力。评估每一个...

Among all of the possible chemical compounds, it’s estimated that between 1020 and 1060 may hold potential as small-molecule drugs. Evaluating each of...

Justin Solomon appointed associate dean of engineering education

在这个新职位, Solomon 将专注于推进整个学校工程教育的创新。他将帮助塑造新的教学方法......

In this new role, Solomon will focus on advancing innovation in engineering education across the school. He will help shape new pedagogical approaches...

Fully Open Meditron: An Auditable Pipeline for Clinical LLMs

临床决策支持系统 (CDSS) 需要可仔细的, 可审计管道，以实现严格的, 可重复验证。然而目前基于法学硕士的 C...

Clinical decision support systems (CDSS) require scrutable, auditable pipelines that enable rigorous, reproducible validation. Yet current LLM-based C...

FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast

LLM 代理能否通过自生成内存而无需梯度更新来改进决策? 我们建议 FORGE (Failure-Optimized Reflective Graduati...

Can LLM agents improve decision-making through self-generated memory without gradient updates? We propose FORGE (Failure-Optimized Reflective Graduati...

Prospective multi-pathogen disease forecasting using autonomous LLM-guided tree search

传染病的概率预测对于公共卫生至关重要，但依赖于专家建模的劳动密集型手动模型管理...

Probabilistic forecasting of infectious diseases is crucial for public health but relies on labor-intensive manual model curation by expert modeling t...

Small, Private Language Models as Teammates for Educational Assessment Design

生成式人工智能越来越多地支持教育设计任务,，例如,通过大型语言模型(LLMs),展示了设计评估的能力...

Generative AI increasingly supports educational design tasks, e.g., through Large Language Models (LLMs), demonstrating the capability to design asses...

Orchard: An Open-Source Agentic Modeling Framework

代理建模旨在将 LLM 转变为能够通过规划, 推理, 工具使用, 和多回合解决复杂任务的自主代理...

Agentic modeling aims to transform LLMs into autonomous agents capable of solving complex tasks through planning, reasoning, tool use, and multi-turn ...