Chips and semiconductors: research

Journal articles, academic blogs, think-tank reports and papers in this section, newest first. Back to the news.

arXiv (research papers) Pengwen Chen

Mixed-Size Placement Initialization via Stiefel-Manifold Sequential Subspace Optimization

Mixed-size global placement simultaneously optimizes the locations of large macros and millions of standard cells in modern very-large-scale integration (VLSI) physical design. State-of-the-art analytical placers achieve strong scalability, but their nonlinear objectives and density handling make them sensitive to the initial placement. This paper presents a geometry-driven framework that formulates placement initialization as a Quadratically Constrained Quadratic Program…

Read at arXiv (research papers)

arXiv (research papers) Zichen Kong

PVTSizing 2.0: A Fast and Reliable Analog Circuit Sizing Operator for Agentic EDA Workflows

Analog circuit design automation requires both coordinated tool execution and efficient numerical search. Large language models (LLMs) offer new ways to organize tool execution, but finding one sizing that satisfies specifications across process, voltage, and temperature (PVT) corners remains costly. PVTSizing 2.0 addresses these needs through a human-agent workflow built around an independently usable numerical optimizer. An agent carries the designer's approved task…

Read at arXiv (research papers)

arXiv (research papers) Yuji Yamamoto Security

From Token Changes to Task Failures: Bit Flips in a Shared KV-Cache Block

In LLM serving systems with prefix caching, each KV block of a shared system prompt is held in GPU memory as a single physical copy, which every request with that prompt reads. Our earlier work measured change and showed that one bit flip in such a block changes many later responses without making them visibly degenerate. This paper extends that work by measuring how often the changed responses are wrong. Using software fault injection into vLLM serving Qwen3-8B, we flip the…

Read at arXiv (research papers)

arXiv (research papers) Xingchen AGI Lab SI

TeleChat to Xing4: Advancing Frontier Agentic Models on Domestic Chips and Frameworks

We present Xing 4, a series of large language models that extends the TeleChat lineage through coordinated advances in model architecture and training infrastructure on Ascend. This report describes the design principles and systems engineering behind the series, spanning architecture selection, large-scale pretraining, and reasoning and agentic post-training. We study how sparse activation, attention design, model depth and width, and speculative decoding balance model…

Read at arXiv (research papers)

arXiv (research papers) He Zhu

Accelerating 3D Gaussian Splatting Training through Selective Freezing and Computation Caching

3D Gaussian Splatting (3DGS) becomes a key primitive for immersive 3D experiences. However, its applicability is limited by its long training overhead. Although prior work has shown that gradient accumulation is the dominant bottleneck in training, existing works either have limited speedup or degrade reconstruction quality. To address this issue, we present ElsaGS, an algorithm-hardware co-designed framework that accelerates 3DGS training with negligible accuracy…

Read at arXiv (research papers)

arXiv (research papers) Muhammed Fatih Balın

Computed Permutation Trellis Coding: Near-Optimal Quantization from Two Multiplications and a 64-Byte Table

We propose Computed Permutation Trellis Coding (CPTC), a low-bit quantization format for tensors written once and read many times. Formats that round each value on its own, such as INT4 and FP4, waste a large part of their bits against the information-theoretic ideal. Trellis codes recover this loss by encoding values jointly, but prior decoders either need a large lookup table for their best quality or compute values with a hash that restricts what they can decode. CPTC's…

Read at arXiv (research papers)

arXiv (research papers) Jacob Z. Williams

Accelerating auxiliary-field quantum Monte Carlo with mixed-precision random propagators

We develop an algorithm to propagate auxiliary-field quantum Monte Carlo (AFQMC) wavefunctions in mixed precision. We target NVIDIA Tensor Cores using the TensorFloat-32 (TF32) format. We approximate each full-precision array by a TF32 leading term and a TF32 residual, and each full-precision matrix product by three TF32 products accumulated in full precision. Benchmarks on hydrogen chains show that mixed-precision propagation is consistently faster than double-precision GPU…

Read at arXiv (research papers)

arXiv (research papers) Varun Darshana Parekh

ThermLeT: Transformer-based Temperature Prediction for 2.5D Chiplet Architecture

The drive toward heterogeneous 2.5D chiplet packages promises unmatched performance and customization, yet the resulting density amplifies thermal challenges. To facilitate rapid design space exploration, we present ThermLeT, a multimodal transformer-based, scale invariant surrogate that integrates power density maps, spatial coordinate embeddings, material properties and heat transfer rate to predict steady state temperature in 13 ms, about four orders of magnitude faster…

Read at arXiv (research papers)

CEPA (Center for European Policy Analysis) Jack Galloway Geopolitics

The Nvidia Scandal Exposes American Blind Spots

A spy scandal underlines the holes in US efforts to keep China from obtaining cutting-edge US technology.

Read at CEPA (Center for European Policy Analysis)

arXiv (research papers) Vijay Nagarajan

Do Apple M-Series Processors Enforce the Compound Memory Model? Sequential and Parallel Composition of Arm and TSO

Apple's M-series processors can run each thread under the Arm memory model or under a TSO model, and threads of both kinds can share memory. This raises two questions of composition that Apple has not answered. Across threads, how does the machine behave as a whole? Within a thread, which Arm instructions play the roles of TSO's loads, stores and atomic read-modify-writes, and how are they ordered between themselves and across other Arm instructions? We propose that both…

Read at arXiv (research papers)

arXiv (research papers) Ben Howe

TCBP-JOBS: GPU-Accelerated Quantum Error Correction Decoding and Noise Learning with Near-Optimal Accuracy

Achieving near-optimal quantum error correction (QEC) decoding accuracy often requires computationally expensive searches that limit both large-scale simulation and real-time operation. We introduce trajectory-calibrated belief propagation (TCBP), which learns calibrated fault probabilities from BP trajectories, and jittered OSD beam search (JOBS), which uses these probabilities to guide parallel GPU searches. Across surface, color and bivariate-bicycle code memory and…

Read at arXiv (research papers)

arXiv (research papers) Lei Tian

GNoF: A Low-SM-Overhead GPU-Native NVMe-over-Fabrics Completion Path for Data-Intensive GPU Workloads

GPUs are the main compute engine for data-intensive workloads whose datasets exceed device memory and reside on remote flash. The conventional host CPU orchestrates every NVMe-over-Fabrics (NVMe-oF) transaction and stages data through host memory, costing high CPU usage, repeated CPU-GPU synchronization, and redundant data movement. GPU-initiated storage removes the CPU from the I/O path, but existing designs leave two problems: synchronous lock-step completion pins…

Read at arXiv (research papers)

arXiv (research papers) Haoran Shang

SURMA: Efficient Fine-Grained Communication Offload for SuperPOD

GPU SuperPODs underpin large language model (LLM) serving, but growing bandwidth and round-trip times make inter-GPU communication consume increasing compute resources. SM-driven remote memory access occupies shared memory or registers for in-flight data and competes for SM-to-NoC bandwidth. Dedicated engines can free these resources, but must handle fine-grained transfers, down to a 576\,B KV-cache entry. NVIDIA's CopyEngine uses less than 1\% of NVLink bandwidth at 7\,KB…

Read at arXiv (research papers)

arXiv (research papers) Bolin Zhang

SCOPE: A Physical Execution Latency-Guided Quantum Circuit Compilation Framework

Quantum computing requires compilers to translate algorithms represented in the quantum circuit model into unitary operations supported by physical devices. Limited coherence time makes physical execution latency an important compilation objective. Gate fusion can shorten pulses, but gate count and depth do not capture fused targets' control demands. Evaluating every candidate with quantum optimal control (QOC) would embed pulse optimization in structural search. We present…

Read at arXiv (research papers)

arXiv (research papers) Wenrui Zhang

QuarkDB: A Hybrid Database System for Quantum Data Management

Emerging quantum applications like quantum communication generate large volumes of high-dimensional quantum data, requiring efficient data analysis and storage. However, classical database systems often struggle to meet requirements such as quantum processing or version management. This paper presents QuarkDB designed for efficient quantum data management. QuarkDB establishes a unified object model for relational and quantum data and defines the syntax of Quantum Structured…

Read at arXiv (research papers)

arXiv (research papers) Zhe Dong

Building Fully On-Device GUI Agents with Mobile NPUs

Mobile GUI agents are increasingly used to automate everyday smartphone tasks, but most existing systems still rely on cloud-hosted models or remote inference, introducing repeated communication latency, privacy exposure, and dependence on network availability. Modern smartphones integrate neural processing units (NPUs) that provide an attractive substrate for fully on-device LLM inference. However, directly deploying conventional GUI-agent pipelines on mobile NPUs is…

Read at arXiv (research papers)

Verfassungsblog Moritz von Rochow Geopolitics

Die Reichweite der europäischen Beistandspflicht

Stellen wir uns einmal vor, Deutschland ist einem Bündnispartner zum Beistand verpflichtet, der Bundestag verweigert aber die vorherige Zustimmung zum militärischen oder zivilen Beistand – kein unwahrscheinliches Szenario. Mit Blick auf die EU hat das Bundesverfassungsgericht festgestellt, dass auch die Europäische Union vertretbar als ein System gegenseitiger kollektiver Sicherheit angesehen werden könne. Wie konkret der wehrverfassungsrechtliche Parlamentsvorbehalt in der…

Read at Verfassungsblog

arXiv (research papers) Ahmed Nematallah

One Mean Can't Rule Them All

There has been extensive debate on how to summarize performance results, especially within the computer architecture community. This article presents a systematic approach to choosing the correct mean, aiming to answer this question for all computer scientists once and for all.

Read at arXiv (research papers)

arXiv (research papers) Linliang Chen SI

GEAR$^2$: Pruned Experts Live On through Gated Reuse of Activated Experts and a Missing-Mass Residual

Mixture-of-Experts (MoE) Large Language Models (LLMs) activate only a few experts per token while generally keeping all the experts in GPU memory for high efficiency. Expert pruning reduces this GPU memory footprint cost, yet discards the contribution of pruned experts. Existing remedies either rely on static substitutes, merge experts at the cost of altering their functions, or require costly recovery training. In this paper, we find that the functionalities of pruned…

Read at arXiv (research papers)

arXiv (research papers) Seohyun Kim

Clock Distribution Network with Backside Clock Mesh in BSPDN Architecture

We propose a clock distribution network (CDN) in which the clock mesh is located on backside metal (BM) layers, while the rest of the network remains on the frontside. Since the backside power distribution network (BSPDN) also uses the BM layers, the clock mesh and power network must share them. We identify an important problem of placing mesh receivers while accounting for keep-out zones (KOZs) between clock- and power-nTSVs. We show that this problem can be solved using a…

Read at arXiv (research papers)

arXiv (research papers) Shiva Raja

Integrated GPU Core for Mamba-3

Mamba-3, the latest Mamba state-space model (SSM), enables efficient long-context modeling, matching Mamba-2 perplexity with half the state size and increasing decoding arithmetic intensity. However, its multi-input multi-output (MIMO) computation, exponential-trapezoidal discretization, and rotary positional embeddings (RoPE) introduce dense, element-wise, and sequential operations. Despite kernel fusion, conventional GPUs execute these operations inefficiently. Our…

Read at arXiv (research papers)

arXiv (research papers) Kazuo Aoyama

Architectural Principles for Parameter-Efficient Unitary Transformations in Coherent Optical Neural Networks

Coherent optical neural networks based on interferometric architectures offer a promising route toward high-throughput, energy-efficient computation by exploiting the multiple degrees of freedom of light, including spatial, temporal, and spectral modes. However, practical implementations are constrained by the number of tunable parameters due to fabrication complexity, device footprint, and control overhead. In such parameter-constrained settings, achieving high expressive…

Read at arXiv (research papers)

arXiv (research papers) Yuhao Liu

LENS: Replacing LayerNorm and Softmax in Transformers with Learnable Normalization States

LayerNorm and attention Softmax are critical operations in Transformers. However, they cannot be efficiently mapped to General Matrix Multiplication (GEMM) computations on hardware. Therefore, modern hardware accelerator designs, such as NPUs on ASIC/FPGA, consume a large amount of hardware resources or chip area to implement their specific arithmetic units. Building on previous work about LayerNorm, Softmax, and their alternatives, we introduce Learnable Normalization…

Read at arXiv (research papers)

arXiv (research papers) Yousef Elbrolosy

feedback-grape: A JAX-based Python Package for GPU-Accelerated GRAPE with Feedback

Modern quantum technology applications require accurate control of quantum systems, including a feedback loop in the control process. While open-loop control can be performed using well established gradient methods like GRadient Ascent Pulse Engineering (GRAPE), it was until recently unclear how to combine these with feedback from measurements, which are inherently non-differentiable. Recent work has successfully demonstrated how feedback can be incorporated into GRAPE using…

Read at arXiv (research papers)

arXiv (research papers) Luciano Bozzi Security

Cyber-Secure Wearable, Ultra-Low-Power Networked Sensors for Soldier Health Monitoring: Insights from the CAPSARII project

The CAPSARII project advances a cyber-secure, ultra-low-power wearable platform and Internet of Battlefield Things (IoBT) framework to support real-time monitoring of military personnel and data-driven decision support. The architecture integrates smart-textile sensing, a body/squad area mesh, edge-deployed AI on secure RISC-V microcontrollers, and cloud/federated analytics for cross-mission learning and comparative studies. Two representative use cases--pilot fatigue and…

Read at arXiv (research papers)

arXiv (research papers) Bingyang Liu

CleanRoute: Reconciling DRC Compliance, Post-Layout Performance, and Runtime in Analog Routing

Automated analog routing remains a long-standing challenge because design-rule checking (DRC) compliance, post-layout performance preservation, and runtime reduction are difficult to achieve together. Existing analog routers have limited coverage of joined-metal design rules, lack control over topology-induced mismatch, and face a trade-off between DRC violations and runtime on uniform routing grids. In this paper, we introduce CleanRoute, an analog router that reconciles…

Read at arXiv (research papers)

arXiv (research papers) Navaneeth Kutty

Exhaustive Evaluation of Fixed-Point Reciprocal Architectures for FPGAs

Fixed-point reciprocals are used in normalization and division on field-programmable gate arrays (FPGAs). Designs that share an output format can differ in worst-case error and in their need for memory and multipliers. The effect of table width, internal precision and refinement was evaluated for reciprocal computation in the Q16.16 and Q8.24 formats. Every representable input in the interval from 0.5 to 2.5 was tested against an exact integer reference. Five arithmetic…

Read at arXiv (research papers)

arXiv (research papers) Achintya Yadav

Does synthetic pre-training pay off on a microcontroller? Accuracy, latency and energy across two MCUs

Rendered images are cheap to label, but a microcontroller classifier must survive 8-bit integer (INT8) quantisation and compute correctly on the chip. Five models, from MobileNetV2 at 224 px to a 61K-parameter convolutional neural network (CNN), were fine-tuned on a six-class scene dataset with and without a synthetic stage, against thresholds fixed in advance, and deployed on an ESP32-S3 and an RP2040. With full real data the stage gave no detectable gain. For the main…

Read at arXiv (research papers)

arXiv (research papers) Eva Sophia Schreyer Security

CHERI under Fault Injection: An Emulation-Based Security Evaluation of Capability Hardware Enhanced RISC Instructions

Memory-safety vulnerabilities continue to dominate real-world exploit chains despite decades of research into software mitigations. CHERI is a hardware architecture that enforces spatial and temporal memory safety by augmenting pointers with hardware-checked metadata, thus preventing software-based exploitation, but its threat model explicitly excludes fault injection. This thesis investigates whether targeted fault injection can undermine CHERI's memory-safety guarantees in…

Read at arXiv (research papers)

arXiv (research papers) Puyang Xu Security

Local Order Matters: A Lightweight CPU-Trainable CNN for Website Fingerprinting on Defended Tor Traffic

Website fingerprinting (WF) attacks let a passive adversary infer which site a Tor user visits from the encrypted cell sequence. On undefended traffic, hand-crafted-feature attacks are already near-ceiling, so deep learning offers only a marginal advantage there. We show that the decisive regime is defended traffic: against the FRONT padding defense, which injects dummy cells that corrupt the volume and count features feature attacks rely on, a compact 95,103-parameter 1-D…

Read at arXiv (research papers)

Older stories