bupt.ai Research & perspectives

AIDC RESEARCH

Computing Research Frontiers.

Research connecting computing, power and thermal management

DAILY RESEARCH EDITIONCooling · Compute–power coordination · AI infrastructure

In focus/ Research highlights

8 papers4 research areas

AI & modeling/ Related research

Beyond Silicon: Materials, Mechanisms, and Methods for Physical Neural Computing

Physical implementations of neural computation now extend far beyond silicon hardware, encompassing substrates such as memristive devices, photonic circuits, mechanical metamaterials, microfluidic networks, chemical reaction systems, and living neural tissue. By exploiting intrinsic physical processes such as charge transport, wave interference, elastic deformation, mass transport, and biochemical regulation, these substrates can realize neural inference and adaptation directly in matter. As silicon GPU- centered

arXivPreprint

READING NOTEStart with the research insight, then examine the methods, comparisons and conditions.

RESEARCH & PERSPECTIVES

Papers & insights

Leave each paper with a question worth pursuing.

8 papers

AI & modeling

03

Beyond Silicon: Materials, Mechanisms, and Methods for Physical Neural Computing

arXivPreprint

Stefan Fischer · Nihat Ay et al.

Research overview

Physical implementations of neural computation now extend far beyond silicon hardware, encompassing substrates such as memristive devices, photonic circuits, mechanical metamaterials, microfluidic networks, chemical reaction systems, and living neural tissue. By exploiting intrinsic physical processes such as charge transport, wave interference, elastic deformation, mass transport, and biochemical regulation, these substrates can realize neural inference and adaptation directly in matter. As silicon GPU- centered

Abstract & detailed commentary

Stefan Fischer, Nihat Ay, Olaf Landsiedel, Esfandiar Mohammadi, Sebastian Otte, Bernd-Christian Renner, Nele Rußwinkel

Chinese abstract

Physical implementations of neural computation now extend far beyond silicon hardware, encompassing substrates such as memristive devices, photonic circuits, mechanical metamaterials, microfluidic networks, chemical reaction systems, and living neural tissue. By exploiting intrinsic physical processes such as charge transport, wave interference, elastic deformation, mass transport, and biochemical regulation, these substrates can realize neural inference and adaptation directly in matter. As silicon GPU- centered

Original paperChinese poster

AI & modeling

06

Beyond Traffic Matrix: DELTA -- A DAG-Aware OCS Logical Topology Optimization for AIDCs

arXivPreprint

Niangen Ye · Jingya Liu et al.

Research overview

The rapid scaling of large language models (LLMs) exacerbates communication bottlenecks in AI

Abstract & detailed commentary

Niangen Ye, Jingya Liu, Guofu Zhu, Weiqiang Sun, Weisheng Hu

Chinese abstract

The rapid scaling of large language models (LLMs) exacerbates communication bottlenecks in AI

Original paperChinese poster

AI & modeling

08

AI Sovereignty as National Learning Capacity: A Human- Centered Learning Mechanics Viewpoint on France, the United States, and China

arXivPreprint

Kim Phuc Tran

Research overview

Artificial Intelligence is often discussed in France in terms of investment, compute capacity, regulation, employment, sovereignty, and education. These dimensions are usually treated separately. This viewpoint paper proposes a unified interpretation: France should be understood as a \emph{national AI

Abstract & detailed commentary

Kim Phuc Tran

Chinese abstract

Artificial Intelligence is often discussed in France in terms of investment, compute capacity, regulation, employment, sovereignty, and education. These dimensions are usually treated separately. This viewpoint paper proposes a unified interpretation: France should be understood as a \emph{national AI

Original paperChinese poster