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) Ziqing Mei

HBF-EventLab: A Discrete-Event Simulator for Studying Host-Device Coordination in High Bandwidth Flash

High Bandwidth Flash (HBF) combines NAND flash density with high-bandwidth accelerator connectivity, offering additional memory capacity for large language model inference. However, limited hardware availability makes it difficult to evaluate how host behavior and device resource constraints jointly affect HBF request processing. We present HBF-EventLab, an open-source, transaction-level discrete-event simulator based on the Open Compute Project HBF specification…

Read at arXiv (research papers)

arXiv (research papers) Chloe Wong

MIAO: Microscaling Intermediate Accumulation Optimization

Microscaling (MX) quantization enables lower-precision large language model (LLM) inference with limited accuracy degradation, reducing memory requirement and multiplication cost. However, despite reduced operand precision, accumulation is often retained at high floating-point precisions, leaving a substantial hardware cost that operand quantization alone does not address. In particular, systolic arrays, as used in tensor accelerators such as Google TPU and AWS Trainium…

Read at arXiv (research papers)

arXiv (research papers) Xuanyi Tan

DT-MATCH: Defect- and Topology-Aware Chiplet Matching for Yield Optimization in Heterogeneous Integration

Heterogeneous integration (HI) and advanced packaging enable chiplet-based system-on-chip (SoC) designs; however, they also introduce new yield challenges arising from manufacturing defects and complex die-to-die (D2D) interconnect constraints. In particular, lane-level defects, limited spares, and strict structural compatibility requirements impede the efficient assembly of functional systems from defective but repairable chiplets. We present DT-MATCH, a defect- and…

Read at arXiv (research papers)

arXiv (research papers) Aritra Bagchi

Rethinking KV Caching for LLM Decode: Free-Space Optical KV Regeneration for Reduction of Memory Traffic

Long-context large language model (LLM) decoding repeatedly reads the historical key-value (KV) cache, making multi-head attention (MHA) memory-bandwidth-bound. Caching layer-input activations instead halves historical-cache storage and traffic, but requires regenerating KV from each cached activation. On digital accelerators such as GPUs, this regeneration takes far longer than fetching the activation from HBM. The KV regeneration overhead outweighs the savings from reduced…

Read at arXiv (research papers)

arXiv (research papers) Mostafa Darvishi SI

From Sensors to Silicon: Causal Cross-Layer Self-Healing Edge AI for Secure and Energy-Adaptive IoT Systems

The migration of artificial intelligence from centralized cloud infrastructure to resource-constrained Internet-of-Things (IoT) devices reduces communication latency and improves data locality, but it also tightly couples sensing quality, inference behavior, microarchitectural activity, hardware reliability, security, and energy management within the same physical node. Existing edge-intelligence solutions predominantly address these dimensions in isolation, making it…

Read at arXiv (research papers)

arXiv (research papers) Spencer H. Bryngelson

A predictive, per-engine energy roofline of the Apple Neural Engine

The Apple Neural Engine (ANE) is in over a billion devices, but is characterized indirectly via CoreML. Using a CoreML-free fused-graph dispatch path, we measure the ANE, along with the Apple GPU and CPU on the same device, with per-engine energy accounting. The consumer-available roofline does not predict the ANE in practice. A critical correction is that the ANE sees two effective memory bandwidths: a slow activation path and a fast weight-stream path. Two roofline ridges…

Read at arXiv (research papers)

arXiv (research papers) Yang Liu

Run, Read Back, Refine: Observation Fidelity in Execution-Grounded EDA Agents

Design engineers lose the most time orchestrating flows and patching the gaps where closed EDA tools refuse to interoperate. We present an execution-grounded LLM agent for PyAether, SKILL, and Tcl that closes a generate-execute-refine loop over unified MCP tools (API retrieval, documentation, sandbox execution) and is already deployed in production EDA flows, with a grounding-gap framework positing that EDA-scripting correctness is bottlenecked by observation fidelity at the…

Read at arXiv (research papers)

arXiv (research papers) Subhadip Ghosh

NEMESIS2.0: LLM-Aided Netlist-Driven Performance Equation Modeling for Accelerated Analog Circuit Sizing

This work presents NEMESIS2.0, an open-source LLM-aided framework that automatically derives equation-based performance models for analog circuits. The resulting equations provide fast, accurate surrogates for SPICE evaluation. To validate the method, equation-based models are developed for seven circuit test cases in a commercial 65nm CMOS technology, covering 15 metrics on five operational transconductance amplifier topologies, and six metrics on two dynamic comparator…

Read at arXiv (research papers)

arXiv (research papers) Darsh Asher Security

Lasso: Towards Sustainable AI-Generated Media Detection with Intel QAT On-Chip Compression

Synthetic text, images, and video are rising. Software-only detectors are power-hungry, too slow, and costly, and do not scale with demand. When software cannot scale, historically hardware often comes to the rescue. We perform the first reverse engineering of a compression accelerator--Intel QuickAssist Technology (QAT) microarchitecture and--find different amount of time on each, a novel data structure-dependent property. Using this information, we then build Lasso, a…

Read at arXiv (research papers)

arXiv (research papers) Valavan Manohararajah

Exstill: The Most Efficient Single-Chip Processor for the Next Token

Generating a token with a language model requires a fixed amount of arithmetic, whose energy sets a floor beneath any hardware. For Qwen3-1.7B at a 16K-token context, gate-level measurements at 5 nm place this floor at 0.89 mJ per token, yet an NVIDIA B300 spends 391-1,120x that. The excess is the price of generality, software layers, programmable hardware and moving weights and key-value (KV) caches from external memory, and it widens with each process generation. We…

Read at arXiv (research papers)

arXiv (research papers) Dayou Du

WaferTetris: Efficient Freeform Mapping on Wafer-Scale Accelerators

Wafer-scale accelerators offer massive compute capacity through hundreds of thousands of processing elements (PEs) interconnected across a single wafer, but realizing this potential requires efficient application mapping. Their scale makes programming these systems and mapping applications onto them challenging. Existing approaches rely on template-based mappings to maintain productivity, but these templates restrict the mapping design space. In particular, they impose fixed…

Read at arXiv (research papers)

arXiv (research papers) Amir Mohammad Khadem Hosseini

EventSpikeSeg: Memory-Aware Mixed-Signal Architecture for Event-Camera Spiking Segmentation

Event-camera segmentation requires persistent neural state despite sparse input events, making memory organization a central constraint on embedded implementation. We present EventSpikeSeg, a mixed-signal architecture combining resident capacitor memory with shared analog processing in GF180MCU 180-nm CMOS. Its 25-population spiking encoder-decoder has 143,824 synaptic weights but requires 30,976,000 membrane states at 440 x 640 resolution. The mapping specifies…

Read at arXiv (research papers)

arXiv (research papers) Yimin Gao

Mutable Inference: Architecting LLM Accelerators for Inference-Time Learning

Long-context LLMs must retain growing histories, but full attention makes each new token increasingly expensive. Inference-time learning (ITL) offers another approach. The model keeps training on the context during inference, retaining what it learns in its weights. Recent work shows that ITL can match full-context attention quality while attending only to a recent fixed-size window. This bounds attention costs but adds Learning (backward computation and weight updates)…

Read at arXiv (research papers)

arXiv (research papers) Qiufeng Li

OpenLayout: An Open and Agentic-Driven Platform for Holistic VLSI Physical Design and Benchmarking

Physical design automation of very large-scale integrated circuits (VLSI) has made significant strides, progressing from single-stage placement and routing toward increasingly integrated full-flow implementation. Yet, existing approaches struggle with the vast design space optimization (DSO) arising from diverse technology nodes, circuit designs, and varied constraints. The substantial runtime of physical-design tools further exacerbates this challenge, leaving industrial…

Read at arXiv (research papers)

arXiv (research papers) Cong Xu

The State Is Not the Bottleneck: A Paged-State Gated-DeltaNet Tile on a Tri-Mode Multicast Mesh, Bit-Exact from RTL to Silicon

Linear-attention language models replace a growing key-value cache with a recurrent state of fixed size. On an edge FPGA, this state can exceed the on-chip memory and must be paged to external memory. This paper asks whether such paging dominates the cost of decoding (generating tokens). We built a hardware tile that computes one complete layer of a custom 369-million-parameter hybrid Gated DeltaNet model on an Agilex 5 FPGA and pages its recurrent state. The model state…

Read at arXiv (research papers)

arXiv (research papers) Jiale Mi

Chip-Based Discrete-Modulated Continuous-Variable Quantum Key Distribution with Uncertainty-Principle-Based Composable Security

Continuous-variable quantum key distribution offers a promising platform for scalable quantum networks through its compatibility with standard telecommunications technology and photonic integration. High-order discrete modulation facilitates practical implementation while approaching the key-rate performance of ideal Gaussian modulation, but its finite-size composable security analysis entails substantial computational overhead. Here, we develop an…

Read at arXiv (research papers)

arXiv (research papers) Jonathan Pan Security

From Bytes to Behavior: Request-Level Attribution from Live GPU Hidden States

Existing memory forensics for machine learning systems typically targets static artifacts, such as model weights or prompt logs, rather than the transient computational states of a live inference process. This gap limits an examiner's ability to analyze the specific behavior of a running large language model (LLM) service that cannot be restarted or modified. We present a method for acquiring per-layer residual-stream activations from a live GPU process using a non-modifying…

Read at arXiv (research papers)

arXiv (research papers) Taro Suzuki

NPU-Accelerated GNSS Signal Acquisition

On small single-board computers (SBCs), GNSS signal acquisition is a major CPU load. This paper proposes using the neural processing unit (NPU) of an SBC not for machine learning but as an accelerator for the parallel code search in GNSS signal acquisition. The discrete Fourier transform (DFT) used in acquisition is decomposed by the four-step DFT into matrix products of small DFTs and expressed by $1\times1$ convolutions, element-wise products, and transposes, which NPUs…

Read at arXiv (research papers)

arXiv (research papers) Marco Giunti Tech policy

The Transformer Paradigm: Dynamic Processing and a New Principle of Inference

We argue that the Transformer revolution is not adequately understood as a merely quantitative expansion of deep learning, but as the emergence of a new paradigm based on a different principle of inference. We interpret the Transformer as a system in which token representations are transformed by parameterized transformations, or transforming concepts. During inference, some of these transformations are dynamically constructed from the current input: previously acquired…

Read at arXiv (research papers)

arXiv (research papers) Kuai Yu

DiffusionSSD: Achieving Extreme Visual Compression with Dual-stage Diffusion and Cross-level Near-Data Processing

Growing visual data overwhelms storage on resource-constrained edge devices, making compression essential. Traditional codecs offer inadequate rate-distortion performance, motivating the exploration of diffusion codecs. Diffusion codecs face two limitations: (1) At the algorithm level, diffusion decoders preserve perceptual quality under extreme compression, but require substantial computation. (2) At the architecture level, in NPU-centered devices, neural codecs occupy the…

Read at arXiv (research papers)

arXiv (research papers) Nikhil Rout

SIMTurn: Lightweight Divergent Subgroup Scheduling for Intra-Warp Synchronization

Stack-based reconvergence enables efficient SIMT execution with a shared warp program counter and compact control-flow state. However, its fixed serialization of divergent paths can deadlock programs that rely on intra-warp synchronization: a spinning subgroup may prevent a resource-holding sibling from executing. Independent Thread Scheduling (ITS) enables finer-grained progress through per-thread execution state, but introduces additional hardware cost. We present SIMTurn…

Read at arXiv (research papers)

arXiv (research papers) Konstantin J Hossfeld

Elasticy: A Runtime-Reconfigurable Architecture for Dynamic Streams

Reconfigurable Dataflow Accelerators (RDAs) have demonstrated high performance and energy efficiency for dense, regular tensor applications. However, conventional RDAs with static configurations show dramatically reduced performance and efficiency on irregular tensor workloads with data-dependent, non-uniform, dynamic tensor shapes. We propose Elasticy, a dynamically reconfigurable dataflow accelerator that employs streaming and tile abstractions to minimize the overhead of…

Read at arXiv (research papers)

arXiv (research papers) Md Omar Faruque

CHARMING: Parallel Edge AI Acceleration Across Tensor Processing and In-Memory Computation

Edge AI increasingly relies on specialized accelerators for low-latency inference under tight resource and energy constraints, yet their efficiency remains limited by data movement. Prior work, CHARM, mitigates this overhead by overlapping data movement with computation. We observe that in-memory computing can further supplement tensor processing as a parallel accelerator while eliminating certain data movements entirely. We present CHARMING, a heterogeneous edge AI…

Read at arXiv (research papers)

arXiv (research papers) Wei Cheng

uhd-rfsoc: Bringing Off-the-Shelf RFSoC Boards to Unmodified UHD Applications

Next-generation wireless research relies on software-defined radios (SDRs) with more channels and larger bandwidth. The USRP family is one of the most widely used SDR platforms, and much of the wireless software ecosystem is built on top of the USRP hardware driver (UHD). However, each USRP model comes with a fixed channel count, bandwidth, and streaming capability. Off-the-shelf radio frequency system-on-chip (RFSoC) boards, on the other hand, integrate up to 16…

Read at arXiv (research papers)

arXiv (research papers) Zhaoqiang Bai

A Unified Device-Level Physical Framework for DRAM Read Disturbance

Read disturbance in dynamic random-access memory (DRAM) depends on both local device interactions and the electrical conditions imposed by shared array circuitry. Here, we combine DIMM-level characterization results, device measurements, and TCAD simulations to interpret RowHammer, RowPress, and ColumnDisturb within a unified device-level physical framework. We relate charge evolution at the floating storage node to three key mechanisms: trap-assisted electron migration…

Read at arXiv (research papers)

Atlantic Council Atlantic Council Geopolitics

What Jensen Huang’s Nvidia exit means for Washington, Wall Street, and Beijing

Our experts unpack what the alleged Nvidia espionage incident means for national security, markets, and relations with Beijing.

Read at Atlantic Council

arXiv (research papers) Haozhe Wang SI

Metal Sandbox: Isolated Execution and Command Scheduling for GPU Coding Agents on Apple Silicon

On-device inference requires kernels that make effective use of local accelerators. Efficient training and evaluation of models that generate Apple Metal kernels benefit from native macOS tools and concurrent agent sessions. These sessions alternate between model requests, development commands, and GPU execution, leaving GPU capacity idle for much of a rollout. We present Metal Sandbox, infrastructure that combines restricted per-instance file access, native command…

Read at arXiv (research papers)

arXiv (research papers) Arnab Sanyal

LazyWeights: Decode-on-Demand Edge LLMs

Memory decides which LLMs an edge device runs, and how fast: capacity on a 4 GB Jetson Nano, bandwidth on an 8 GB Jetson Orin Nano Super. Deployed formats (GPTQ, AWQ, llama.cpp's K- and I-quants) spend equal bits on every integer code of a matrix, although the codes are far from uniform. LazyWeights keeps these formats but tunes GPTQ so the codes compress, Huffman-codes them losslessly in interleaved lanes, and decodes them on demand in a small unit between DRAM and L2, so…

Read at arXiv (research papers)

arXiv (research papers) Hannah Atmer

Broadcast, Don't Replicate: Eliminating Weight HBM in Datacenter LLM Serving

Large language model (LLM) inference at datacenter scale is fundamentally bottlenecked by memory capacity and bandwidth limits. Because model weights are read-only and identical across instances, conventional architectures that replicate these parameters across the local DRAM of tens of thousands of accelerators incur massive capital expenditures, significant static power dissipation, and redundant memory-fetch energy. A recent proposal to use optical fiber loops as…

Read at arXiv (research papers)

Older stories