启用隐私
Enabling privacy
麻省理工学院研究人员开发的一种新方法可以将保护隐私的人工智能训练方法加速约 81%。这种先进...
A new method developed by MIT researchers can accelerate a privacy-preserving artificial intelligence training method by about 81 percent. This advanc...
01
The MIT
The MIT
IBM 和麻省理工学院今天宣布成立 MIT-IBM 计算研究实验室,，以推进他们的长期合作，以塑造下一个计算机时代。
IBM and MIT today announced the launch of the MIT-IBM Computing Research Lab, advancing their long-standing collaboration to shape the next era of com...
02
激发好奇心
Making the case for curiosity
“当我来到麻省理工学院时，真正让我震惊的事情，并且每天都让我震惊的是， 这里发生的事情是令人惊奇的。科学, 工程...
“The thing that really struck me when I came to MIT and strikes me every single day is the stuff that’s going on here is amazing. The science, the eng...
03
解决 “Whac
Solving the “Whac
在当今’家医院和诊所,，皮肤科医生可能会使用人工智能模型对皮肤病变进行分类，以评估病变是否处于...
In today’s hospitals and clinics, a dermatologist may use an artificial intelligence model for classifying skin lesions to assess if the lesion is at ...
04
治理您无法观察到的内容: 自主 AI 代理的自适应运行时治理
Governing What You Cannot Observe: Adaptive Runtime Governance for Autonomous AI Agents
自主人工智能代理可以保持完全授权，但随着行为漂移,对手适应,并且决策模式在没有...的情况下发生变化，仍然变得不安全。
Autonomous AI agents can remain fully authorized and still become unsafe as behavior drifts, adversaries adapt, and decision patterns shift without an...
05
当前代理能否缩小发现到应用程序的差距? Minecraft 中的案例研究
Can Current Agents Close the Discovery-to-Application Gap? A Case Study in Minecraft
发现因果规律并将其应用于构建功能系统（发现到应用循环）是通用智能的标志...
Discovering causal regularities and applying them to build functional systems--the discovery-to-application loop--is a hallmark of general intelligenc...
06
临床 AI 评估的具体案例评估: 方法, 验证, 和法学硕士-临床医生在 823 次接触中达成的协议
Case-Specific Rubrics for Clinical AI Evaluation: Methodology, Validation, and LLM-Clinician Agreement Across 823 Encounters
客观的。临床 AI 文档系统需要临床有效, 经济可行, 且对迭代敏感的评估方法...
Objective. Clinical AI documentation systems require evaluation methodologies that are clinically valid, economically viable, and sensitive to iterati...
07
学习旋转: 时间和语义旋转编码以进行顺序建模
Learning to Rotate: Temporal and Semantic Rotary Encoding for Sequential Modeling
每个 Transformer 架构都致力于学习语义嵌入空间中丰富的表示的巨大能力 - 然而旋转流形...
Every Transformer architecture dedicates enormous capacity to learning rich representations in semantic embedding space -- yet the rotation manifold a...
08
阅读论文, 编写代码: 社会科学结果的代理再现
Read the Paper, Write the Code: Agentic Reproduction of Social-Science Results
最近的工作使用 LLM 代理来重现实证社会科学结果，并可访问数据和代码。我们通过询问: 来扩大这个范围...
Recent work has used LLM agents to reproduce empirical social science results with access to both the data and code. We broaden this scope by asking: ...
09
重新思考 Publication: AI 支持研究的认证框架
Rethinking Publication: A Certification Framework for AI-Enabled Research
人工智能研究渠道现在可以产生可以满足现有同行评审质量,新颖性,和方法严谨性标准的学术工作。哈...
AI research pipelines can now generate academic work that may satisfy existing peer review standards for quality, novelty, and methodological rigor. H...
10