BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Minyi Guo

Minyi Guo contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Machine LearningDistributed, Parallel, and Cluster ComputingArtificial IntelligenceComputation and LanguageHardware ArchitectureCryptography and SecurityDatabaseseess.SPProgramming Languages

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
15works
0followers
9topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

15 published item(s)

preprint2026arXiv

CuBridge: An LLM-Based Framework for Understanding and Reconstructing High-Performance Attention Kernels

Efficient CUDA implementations of attention mechanisms are critical to modern deep learning systems, yet supporting diverse and evolving attention variants remains challenging. Existing frameworks and compilers trade performance for flexibility, while expert-written kernels achieve high efficiency but are difficult to adapt. Recent work explores large language models (LLMs) for GPU kernel generation, but prior studies report unstable correctness and significant performance gaps for complex operators such as attention. We present CuBridge, an LLM-based framework that adapts expert-written attention kernels through a structured lift-transfer-lower workflow. CuBridge starts from expert-written CUDA attention kernels and lifts them into an executable intermediate representation that makes execution orchestration explicit while abstracting low-level CUDA syntax. Given a user-provided PyTorch specification, CuBridge generates and verifies a target IR program, then reconstructs optimized CUDA code via reference-guided lowering. Across diverse attention variants and GPU platforms, CuBridge consistently produces correct kernels and substantially outperforms general frameworks, compiler-based approaches, and prior LLM-based methods.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

On the (In-)Security of the Shuffling Defense in the Transformer Secure Inference

For Transformer models, cryptographically secure inference ensures that the client learns only the final output, while the server learns nothing about the client's input. However, securely computing nonlinear layers remains a major efficiency bottleneck due to the substantial communication rounds and data transmission required. To address this issue, prior works reveal intermediate activations to the client, allowing nonlinear operations to be computed in plaintext. Although this approach significantly improves efficiency, exposing activations enables adversaries to extract model weights. To mitigate this risk, existing works employ a shuffling defense that reveals only randomly permuted activations to the client. In this work, we show that the shuffling defense is not as robust as previously claimed. We propose an attack that aligns differently shuffled activations to a common permutation and subsequently exploits them to extract model weights. Experiments on Pythia-70m and GPT-2 demonstrate that the proposed attack can align shuffled activations with mean squared errors ranging from $10^{-9}$ to $10^{-6}$. With a query cost of approximately \$1, the adversary can recover model weights with L1-norm differences ranging from $10^{-4}$ to $10^{-2}$ compared to the oracle weights.

0 citations0 reviewsNo code linkedOpen access
preprint2025arXiv

Yggdrasil: Bridging Dynamic Speculation and Static Runtime for Latency-Optimal Tree-Based LLM Decoding

Speculative decoding improves LLM inference by generating and verifying multiple tokens in parallel, but existing systems suffer from suboptimal performance due to a mismatch between dynamic speculation and static runtime assumptions. We present Yggdrasil, a co-designed system that enables latency-optimal speculative decoding through context-aware tree drafting and compiler-friendly execution. Yggdrasil introduces an equal-growth tree structure for static graph compatibility, a latency-aware optimization objective for draft selection, and stage-based scheduling to reduce overhead. Yggdrasil supports unmodified LLMs and achieves up to $3.98\times$ speedup over state-of-the-art baselines across multiple hardware setups.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Async-fork: Mitigating Query Latency Spikes Incurred by the Fork-based Snapshot Mechanism from the OS Level

In-memory key-value stores (IMKVSes) serve many online applications because of their efficiency. To support data backup, popular industrial IMKVSes periodically take a point-in-time snapshot of the in-memory data with the system call fork. However, this mechanism can result in latency spikes for queries arriving during the snapshot period because fork leads the engine into the kernel mode in which the engine is out-of-service for queries. In contrast to existing research focusing on optimizing snapshot algorithms, we optimize the fork operation to address the latency spikes problem from the operating system (OS) level, while keeping the data persistent mechanism in IMKVSes unchanged. Specifically, we first conduct an in-depth study to reveal the impact of the fork operation as well as the optimization techniques on query latency. Based on findings in the study, we propose Async-fork to offload the work of copying the page table from the engine (the parent process) to the child process as copying the page table dominates the execution time of fork. To keep data consistent between the parent and the child, we design the proactive synchronization strategy. Async-fork is implemented in the Linux kernel and deployed into the online Redis database in public clouds. Our experiment results show that compared with the default fork method in OS, Async-fork reduces the tail latency of queries arriving during the snapshot period by 81.76% on an 8GB instance and 99.84% on a 64GB instance.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

ANT: Exploiting Adaptive Numerical Data Type for Low-bit Deep Neural Network Quantization

Quantization is a technique to reduce the computation and memory cost of DNN models, which are getting increasingly large. Existing quantization solutions use fixed-point integer or floating-point types, which have limited benefits, as both require more bits to maintain the accuracy of original models. On the other hand, variable-length quantization uses low-bit quantization for normal values and high-precision for a fraction of outlier values. Even though this line of work brings algorithmic benefits, it also introduces significant hardware overheads due to variable-length encoding and decoding. In this work, we propose a fixed-length adaptive numerical data type called ANT to achieve low-bit quantization with tiny hardware overheads. Our data type ANT leverages two key innovations to exploit the intra-tensor and inter-tensor adaptive opportunities in DNN models. First, we propose a particular data type, flint, that combines the advantages of float and int for adapting to the importance of different values within a tensor. Second, we propose an adaptive framework that selects the best type for each tensor according to its distribution characteristics. We design a unified processing element architecture for ANT and show its ease of integration with existing DNN accelerators. Our design results in 2.8$\times$ speedup and 2.5$\times$ energy efficiency improvement over the state-of-the-art quantization accelerators.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Block-Skim: Efficient Question Answering for Transformer

Transformer models have achieved promising results on natural language processing (NLP) tasks including extractive question answering (QA). Common Transformer encoders used in NLP tasks process the hidden states of all input tokens in the context paragraph throughout all layers. However, different from other tasks such as sequence classification, answering the raised question does not necessarily need all the tokens in the context paragraph. Following this motivation, we propose Block-skim, which learns to skim unnecessary context in higher hidden layers to improve and accelerate the Transformer performance. The key idea of Block-Skim is to identify the context that must be further processed and those that could be safely discarded early on during inference. Critically, we find that such information could be sufficiently derived from the self-attention weights inside the Transformer model. We further prune the hidden states corresponding to the unnecessary positions early in lower layers, achieving significant inference-time speedup. To our surprise, we observe that models pruned in this way outperform their full-size counterparts. Block-Skim improves QA models' accuracy on different datasets and achieves 3 times speedup on BERT-base model.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

SALO: An Efficient Spatial Accelerator Enabling Hybrid Sparse Attention Mechanisms for Long Sequences

The attention mechanisms of transformers effectively extract pertinent information from the input sequence. However, the quadratic complexity of self-attention w.r.t the sequence length incurs heavy computational and memory burdens, especially for tasks with long sequences. Existing accelerators face performance degradation in these tasks. To this end, we propose SALO to enable hybrid sparse attention mechanisms for long sequences. SALO contains a data scheduler to map hybrid sparse attention patterns onto hardware and a spatial accelerator to perform the efficient attention computation. We show that SALO achieves 17.66x and 89.33x speedup on average compared to GPU and CPU implementations, respectively, on typical workloads, i.e., Longformer and ViL.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

SQuant: On-the-Fly Data-Free Quantization via Diagonal Hessian Approximation

Quantization of deep neural networks (DNN) has been proven effective for compressing and accelerating DNN models. Data-free quantization (DFQ) is a promising approach without the original datasets under privacy-sensitive and confidential scenarios. However, current DFQ solutions degrade accuracy, need synthetic data to calibrate networks, and are time-consuming and costly. This paper proposes an on-the-fly DFQ framework with sub-second quantization time, called SQuant, which can quantize networks on inference-only devices with low computation and memory requirements. With the theoretical analysis of the second-order information of DNN task loss, we decompose and approximate the Hessian-based optimization objective into three diagonal sub-items, which have different areas corresponding to three dimensions of weight tensor: element-wise, kernel-wise, and output channel-wise. Then, we progressively compose sub-items and propose a novel data-free optimization objective in the discrete domain, minimizing Constrained Absolute Sum of Error (or CASE in short), which surprisingly does not need any dataset and is even not aware of network architecture. We also design an efficient algorithm without back-propagation to further reduce the computation complexity of the objective solver. Finally, without fine-tuning and synthetic datasets, SQuant accelerates the data-free quantization process to a sub-second level with >30% accuracy improvement over the existing data-free post-training quantization works, with the evaluated models under 4-bit quantization. We have open-sourced the SQuant framework at https://github.com/clevercool/SQuant.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

The Serverless Computing Survey: A Technical Primer for Design Architecture

The development of cloud infrastructures inspires the emergence of cloud-native computing. As the most promising architecture for deploying microservices, serverless computing has recently attracted more and more attention in both industry and academia. Due to its inherent scalability and flexibility, serverless computing becomes attractive and more pervasive for ever-growing Internet services. Despite the momentum in the cloud-native community, the existing challenges and compromises still wait for more advanced research and solutions to further explore the potentials of the serverless computing model. As a contribution to this knowledge, this article surveys and elaborates the research domains in the serverless context by decoupling the architecture into four stack layers: Virtualization, Encapsule, System Orchestration, and System Coordination. Inspired by the security model, we highlight the key implications and limitations of these works in each layer, and make suggestions for potential challenges to the field of future serverless computing.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Transkimmer: Transformer Learns to Layer-wise Skim

Transformer architecture has become the de-facto model for many machine learning tasks from natural language processing and computer vision. As such, improving its computational efficiency becomes paramount. One of the major computational inefficiency of Transformer-based models is that they spend the identical amount of computation throughout all layers. Prior works have proposed to augment the Transformer model with the capability of skimming tokens to improve its computational efficiency. However, they suffer from not having effectual and end-to-end optimization of the discrete skimming predictor. To address the above limitations, we propose the Transkimmer architecture, which learns to identify hidden state tokens that are not required by each layer. The skimmed tokens are then forwarded directly to the final output, thus reducing the computation of the successive layers. The key idea in Transkimmer is to add a parameterized predictor before each layer that learns to make the skimming decision. We also propose to adopt reparameterization trick and add skim loss for the end-to-end training of Transkimmer. Transkimmer achieves 10.97x average speedup on GLUE benchmark compared with vanilla BERT-base baseline with less than 1% accuracy degradation.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

VELTAIR: Towards High-Performance Multi-tenant Deep Learning Services via Adaptive Compilation and Scheduling

Deep learning (DL) models have achieved great success in many application domains. As such, many industrial companies such as Google and Facebook have acknowledged the importance of multi-tenant DL services. Although the multi-tenant service has been studied in conventional workloads, it is not been deeply studied on deep learning service, especially on general-purpose hardware. In this work, we systematically analyze the opportunities and challenges of providing multi-tenant deep learning services on the general-purpose CPU architecture from the aspects of scheduling granularity and code generation. We propose an adaptive granularity scheduling scheme to both guarantee resource usage efficiency and reduce the scheduling conflict rate. We also propose an adaptive compilation strategy, by which we can dynamically and intelligently pick a program with proper exclusive and shared resource usage to reduce overall interference-induced performance loss. Compared to the existing works, our design can serve more requests under the same QoS target in various scenarios (e.g., +71%, +62%, +45% for light, medium, and heavy workloads, respectively), and reduce the averaged query latency by 50%.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise Sparsity

Network pruning can reduce the high computation cost of deep neural network (DNN) models. However, to maintain their accuracies, sparse models often carry randomly-distributed weights, leading to irregular computations. Consequently, sparse models cannot achieve meaningful speedup on commodity hardware (e.g., GPU) built for dense matrix computations. As such, prior works usually modify or design completely new sparsity-optimized architectures for exploiting sparsity. We propose an algorithm-software co-designed pruning method that achieves latency speedups on existing dense architectures. Our work builds upon the insight that the matrix multiplication generally breaks the large matrix into multiple smaller tiles for parallel execution. We propose a tiling-friendly "tile-wise" sparsity pattern, which maintains a regular pattern at the tile level for efficient execution but allows for irregular, arbitrary pruning at the global scale to maintain the high accuracy. We implement and evaluate the sparsity pattern on GPU tensor core, achieving a 1.95x speedup over the dense model.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Balancing Efficiency and Flexibility for DNN Acceleration via Temporal GPU-Systolic Array Integration

The research interest in specialized hardware accelerators for deep neural networks (DNN) spikes recently owing to their superior performance and efficiency. However, today's DNN accelerators primarily focus on accelerating specific "kernels" such as convolution and matrix multiplication, which are vital but only part of an end-to-end DNN-enabled application. Meaningful speedups over the entire application often require supporting computations that are, while massively parallel, ill-suited to DNN accelerators. Integrating a general-purpose processor such as a CPU or a GPU incurs significant data movement overhead and leads to resource under-utilization on the DNN accelerators. We propose Simultaneous Multi-mode Architecture (SMA), a novel architecture design and execution model that offers general-purpose programmability on DNN accelerators in order to accelerate end-to-end applications. The key to SMA is the temporal integration of the systolic execution model with the GPU-like SIMD execution model. The SMA exploits the common components shared between the systolic-array accelerator and the GPU, and provides lightweight reconfiguration capability to switch between the two modes in-situ. The SMA achieves up to 63% performance improvement while consuming 23% less energy than the baseline Volta architecture with TensorCore.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Ptolemy: Architecture Support for Robust Deep Learning

Deep learning is vulnerable to adversarial attacks, where carefully-crafted input perturbations could mislead a well-trained Deep Neural Network to produce incorrect results. Today's countermeasures to adversarial attacks either do not have capability to detect adversarial samples at inference time, or introduce prohibitively high overhead to be practical at inference time. We propose Ptolemy, an algorithm-architecture co-designed system that detects adversarial attacks at inference time with low overhead and high accuracy.We exploit the synergies between DNN inference and imperative program execution: an input to a DNN uniquely activates a set of neurons that contribute significantly to the inference output, analogous to the sequence of basic blocks exercised by an input in a conventional program. Critically, we observe that adversarial samples tend to activate distinctive paths from those of benign inputs. Leveraging this insight, we propose an adversarial sample detection framework, which uses canary paths generated from offline profiling to detect adversarial samples at runtime. The Ptolemy compiler along with the co-designed hardware enable efficient execution by exploiting the unique algorithmic characteristics. Extensive evaluations show that Ptolemy achieves higher or similar adversarial example detection accuracy than today's mechanisms with a much lower runtime (as low as 2%) overhead.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Towards QoS-Aware and Resource-Efficient GPU Microservices Based on Spatial Multitasking GPUs In Datacenters

While prior researches focus on CPU-based microservices, they are not applicable for GPU-based microservices due to the different contention patterns. It is challenging to optimize the resource utilization while guaranteeing the QoS for GPU microservices. We find that the overhead is caused by inter microservice communication, GPU resource contention and imbalanced throughput within microservice pipeline. We propose Camelot, a runtime system that manages GPU micorservices considering the above factors. In Camelot, a global memory-based communication mechanism enables onsite data sharing that significantly reduces the end-to-end latencies of user queries. We also propose two contention aware resource allocation policies that either maximize the peak supported service load or minimize the resource usage at low load while ensuring the required QoS. The two policies consider the microservice pipeline effect and the runtime GPU resource contention when allocating resources for the microservices. Compared with state-of-the-art work, Camelot increases the supported peak load by up to 64.5% with limited GPUs, and reduces 35% resource usage at low load while achieving the desired 99%-ile latency target.

0 citations0 reviewsNo code linkedOpen access