News
The Shape Of Prompts: Exploring Their Effect On Inference Infrastructure
1+ hour, 23+ min ago (344+ words) Aligning GPU, memory, storage, and network resources in a balanced and efficient configuration. If prompts are fluid, they are not random. They have structure. And that structure can be visualized. These shapes extend toward the'Latency'axis. The system may not be…...
Foundation Model For Physics: The Next Layer Of Intelligence For Engineering
1+ hour, 20+ min ago (939+ words) Over the past decade or so, foundation models have emerged as the dominant paradigm for interacting with language, images, and code. Large Language Models (LLMs) can generate text. Vision models can interpret images. Multimodal systems can connect the two seamlessly....
Beyond Ideal Crystals: The Case For Scale In Atomistic Modeling
1+ week, 1+ hour ago (446+ words) Almost all computer simulations face the same trade-off: larger models can be more realistic and therefore more useful, but they also take longer to run. Engineers and scientists are therefore faced with an almost daily challenge of choosing a model…...
Four-Tier Memory Hierarchy for LLM Reasoning (USC, UW)
1+ week, 14+ hour ago (371+ words) A new technical paper, "Not All Thoughts Need HBM: Semantics-Aware Memory Hierarchy for LLM Reasoning," was published by researchers at USC and University of Wisconsin-Madison. Abstract "Reasoning LLMs produce thousands of chain-of-thought tokens whose KV cache must reside in scarce…...
Scalable Photomask Optimization With Morphological Learning (SUNY Buffalo, VU, IBM)
1+ week, 2+ day ago (119+ words) Semiconductor Engineering A new technical paper, "Morph OPC: Advancing Mask Optimization with Multi-scale Hierarchical Morphological Learning," was published by researchers at University at Buffalo, Villanova University, and IBM T. J. Watson Research Center. Hu, Yuting, Lei Zhuang, Chen Wang, Ruiyang Qin, Hua…...
HW-Native, GPU Compiler for Large-scale ML Production Systems (UC San Diego, Meta)
1+ week, 2+ day ago (301+ words) A new technical paper, "TLX: Hardware-Native, Evolvable MIMW GPU Compiler for Large-scale Production Environments," was published by researchers at UC San Diego and Meta. Abstract "Modern GPUs increasingly rely on specialized hardware units and asynchronous coordination mechanisms, so performance depends…...
vision-language-action models Archives
2+ week, 41+ min ago (79+ words) Semiconductor Engineering When Semiconductor Materials Misbehave AI Growing Impact On Chip Design And EDA Tools Panel-Level Packaging's Second Wave Meets Engineering Reality TSV Complexity Leads To Manufacturing Bottleneck Liquid Cooling Gains Traction In Data Centers Will 2026 Be Dominated By AI?...
Vision-Language-Action Models Arrive
2+ week, 1+ hour ago (687+ words) An emerging AI architecture for embedded autonomy challenges edge efficiency. This post covers what VLAs are, how they're built, where they become hard to deploy on embedded hardware, and what it takes to run one efficiently at the edge. A…...
HW-Based Image Generation Using FTJs (SNU, Sungkyunkwan U. , SK hynix et al.)
2+ week, 1+ day ago (325+ words) A new technical paper, "CMOS-compatible ferroelectric tunnel junctions integrate stochastic sampling and deterministic computing for image generation," was published by researchers at Seoul National University, Sungkyunkwan University, Hanyang University, Sogang University, and SK Hynix. Abstract "Recent progress in generative modeling…...
Test Distribution Evolves To Meet AI Challenges
2+ week, 2+ day ago (250+ words) ATE is evolving from a pure defect-detection system to one that provides system-level validation supported by AI software tools. The growing complexity of AI-focused semiconductor devices presents major challenges for semiconductor test operations. Transistor counts continue to rise steeply, leading…...