arXiv subjects

Shuwen Deng

Publications and source records attributed to Shuwen Deng.

12 recordsLinked to original sources

arXiv2026

Application-Driven Architecture Exploration for Cross-Layer Heterogeneous Systems

AI and HPC infrastructure increasingly serves workload portfolios that combine dense tensor computation, sparse kernels, large memory footprints, and communication-intensive collectives. Supporting these portfolios requires coordinated choices across accelerators, memory tiers, scale-up fabrics, and cluster networks. The resulting Cross-layer Heterogeneous System (XHS) design space is difficult to explore: hardware choices change legal task mappings, while rack power, switch radix, cabling, and cost constraints invalidate many candidates. We present CHASE, an application-driven framework that searches physically feasible XHS architectures through the workloads they must execute. CHASE represents candidates as hierarchical typed graphs and rejects designs that violate deployment constraints. It avoids intractable joint hardware-mapping search with a decoupled two-level loop: an inner mapper translates hardware-independent workload DAGs into topology-aware event traces, a calibrated event-driven simulator evaluates each mapping, and an outer telemetry-guided optimizer evolves the hardware graph. We evaluate CHASE on sparse-computing and LLM workloads. Its mapper remains within 6.06% of exhaustive optima while reducing mapping time by 60.5% on average relative to PEFT. Compute-model errors average 4.4-7.5%, and communication validation reproduces key trends across physical platforms. The outer search reaches near-global optima within 64 iterations. End-to-end case studies show that sparse workloads favor criticality-aware heterogeneous pods, whereas LLM inference favors scale-up islands; the resulting designs deliver 6.20$\times$ and 2.12$\times$ geomean speedups, respectively, while reducing cost and power relative to the baselines.

cs.DC↗

arXiv2026

Eyettention II: A Dual-Sequence Architecture for Modeling Fixation Location, Within-Word Landing Position, and Fixation Duration in Reading

The way our eyes move while reading provides valuable insights into both the reader's cognitive processes and the properties of the text. In particular, eye-tracking-while-reading data has shown to be highly beneficial in various technological applications, such as enhancing and interpreting language models and inferring a reader's characteristics. However, these applications often rely on large-scale, data-driven models, which demand extensive eye-tracking datasets that are challenging to obtain due to the resource-intensive nature of data collection. To address the challenge of data scarcity, we develop Eyettention II, an end-to-end trained deep-learning model capable of generating realistic scanpaths consisting of a complete set of fixation attributes in chronological order, including fixation location, within-word landing position, and fixation duration. Our model is lightweight, efficiently trainable on limited GPU resources, and closely aligned with cognitive theories. We demonstrate that Eyettention II surpasses state-of-the-art models in scanpath prediction and mirrors human-like gaze behavior by capturing key psycholinguistic phenomena. With its robust performance, Eyettention II holds the potential to drive advancements in natural language processing, facilitate piloting the materials of psycholinguistic experiments, and uncover new insights beyond what is explicitly encoded in theoretical cognitive models.

cs.CL↗

arXiv2024

CXL-Interference: Analysis and Characterization in Modern Computer Systems

Compute Express Link (CXL) is a promising technology that addresses memory and storage challenges. Despite its advantages, CXL faces performance threats from external interference when co-existing with current memory and storage systems. This interference is under-explored in existing research. To address this, we develop CXL-Interplay, systematically characterizing and analyzing interference from memory and storage systems. To the best of our knowledge, we are the first to characterize CXL interference on real CXL hardware. We also provide reverse-reasoning analysis with performance counters and kernel functions. In the end, we propose and evaluate mitigating solutions.

cs.AR↗

arXiv2024

Palermo: Improving the Performance of Oblivious Memory using Protocol-Hardware Co-Design

Oblivious RAM (ORAM) hides the memory access patterns, enhancing data privacy by preventing attackers from discovering sensitive information based on the sequence of memory accesses. The performance of ORAM is often limited by its inherent trade-off between security and efficiency, as concealing memory access patterns imposes significant computational and memory overhead. While prior works focus on improving the ORAM performance by prefetching and eliminating ORAM requests, we find that their performance is very sensitive to workload locality behavior and incurs additional management overhead caused by the ORAM stash pressure. This paper presents Palermo: a protocol-hardware co-design to improve ORAM performance. The key observation in Palermo is that classical ORAM protocols enforce restrictive dependencies between memory operations that result in low memory bandwidth utilization. Palermo introduces a new protocol that overlaps large portions of memory operations, within a single and between multiple ORAM requests, without breaking correctness and security guarantees. Subsequently, we propose an ORAM controller architecture that executes the proposed protocol to service ORAM requests. The hardware is responsible for concurrently issuing memory requests as well as imposing the necessary dependencies to ensure a consistent view of the ORAM tree across requests. Using a rich workload mix, we demonstrate that Palermo outperforms the RingORAM baseline by 2.8x, on average, incurring a negligible area overhead of 5.78mm^2 (less than 2% in 12th generation Intel CPU after technology scaling) and 2.14W without sacrificing security. We further show that Palermo also outperforms the state-of-the-art works PageORAM, PrORAM, and IR-ORAM.

cs.CR↗

arXiv2024

Hecaton: Training Large Language Models with Scalable Chiplet Systems

Large Language Models (LLMs) have achieved remarkable success in various fields, but their training and finetuning require massive computation and memory, necessitating parallelism which introduces heavy communication overheads. Driven by advances in packaging, the chiplet architecture emerges as a potential solution, as it can integrate computing power, as well as utilize on-package links with better signal integrity, higher bandwidth, and lower energy consumption. However, most existing chiplet-related works focus on DNN inference. Directly porting them to LLM training introduces significantly large quantities of DRAM access and network-on-package (NoP) overheads which make state-of-the-art chiplet designs fail, highlighting a research gap. This work proposes Hecaton, a scalable and cost-effective chiplet system for LLM training. We first provide a chiplet architecture with tailored scheduling that can largely reduce DRAM accesses. We further design an efficient distributed training method that reduces NoP communication complexity and relieves constraints on SRAM capacity and layout. Theoretical analysis shows that the entire system achieves weak scaling: as the workload and hardware resources grow proportionally, the computation-to-communication ratio remains nearly constant. Experiments with various workloads and hardware configurations verify the property, and Hecaton achieves $5.29\times$ performance improvement and $3.46\times$ energy reduction on Llama3.1-405B, compared to the tensor parallelism in Megatron. To the best of our knowledge, we propose the first chiplet architecture specifically used for LLM training or finetuning, with guaranteed performance regardless of the problem scale.

cs.AR↗

arXiv2023

Pre-Trained Language Models Augmented with Synthetic Scanpaths for Natural Language Understanding

Human gaze data offer cognitive information that reflects natural language comprehension. Indeed, augmenting language models with human scanpaths has proven beneficial for a range of NLP tasks, including language understanding. However, the applicability of this approach is hampered because the abundance of text corpora is contrasted by a scarcity of gaze data. Although models for the generation of human-like scanpaths during reading have been developed, the potential of synthetic gaze data across NLP tasks remains largely unexplored. We develop a model that integrates synthetic scanpath generation with a scanpath-augmented language model, eliminating the need for human gaze data. Since the model's error gradient can be propagated throughout all parts of the model, the scanpath generator can be fine-tuned to downstream tasks. We find that the proposed model not only outperforms the underlying language model, but achieves a performance that is comparable to a language model augmented with real human gaze data. Our code is publicly available.

cs.CL↗

arXiv2023

Eyettention: An Attention-based Dual-Sequence Model for Predicting Human Scanpaths during Reading

Eye movements during reading offer insights into both the reader's cognitive processes and the characteristics of the text that is being read. Hence, the analysis of scanpaths in reading have attracted increasing attention across fields, ranging from cognitive science over linguistics to computer science. In particular, eye-tracking-while-reading data has been argued to bear the potential to make machine-learning-based language models exhibit a more human-like linguistic behavior. However, one of the main challenges in modeling human scanpaths in reading is their dual-sequence nature: the words are ordered following the grammatical rules of the language, whereas the fixations are chronologically ordered. As humans do not strictly read from left-to-right, but rather skip or refixate words and regress to previous words, the alignment of the linguistic and the temporal sequence is non-trivial. In this paper, we develop Eyettention, the first dual-sequence model that simultaneously processes the sequence of words and the chronological sequence of fixations. The alignment of the two sequences is achieved by a cross-sequence attention mechanism. We show that Eyettention outperforms state-of-the-art models in predicting scanpaths. We provide an extensive within- and across-data set evaluation on different languages. An ablation study and qualitative analysis support an in-depth understanding of the model's behavior.

cs.CL↗

arXiv2022

Detection of ADHD based on Eye Movements during Natural Viewing

Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder that is highly prevalent and requires clinical specialists to diagnose. It is known that an individual's viewing behavior, reflected in their eye movements, is directly related to attentional mechanisms and higher-order cognitive processes. We therefore explore whether ADHD can be detected based on recorded eye movements together with information about the video stimulus in a free-viewing task. To this end, we develop an end-to-end deep learning-based sequence model which we pre-train on a related task for which more data are available. We find that the method is in fact able to detect ADHD and outperforms relevant baselines. We investigate the relevance of the input features in an ablation study. Interestingly, we find that the model's performance is closely related to the content of the video, which provides insights for future experimental designs.

cs.CV↗

arXiv2022

Short Paper: Device- and Locality-Specific Fingerprinting of Shared NISQ Quantum Computers

Fingerprinting of quantum computer devices is a new threat that poses a challenge to shared, cloud-based quantum computers. Fingerprinting can allow adversaries to map quantum computer infrastructures, uniquely identify cloud-based devices which otherwise have no public identifiers, and it can assist other adversarial attacks. This work shows idle tomography-based fingerprinting method based on crosstalk-induced errors in NISQ quantum computers. The device- and locality-specific fingerprinting results show prediction accuracy values of $99.1\%$ and $95.3\%$, respectively.

cs.CR↗

arXiv2022

Leaky Frontends: Security Vulnerabilities in Processor Frontends

This paper evaluates new security threats due to the processor frontend in modern Intel processors. The root causes of the security threats are the multiple paths in the processor frontend that the micro-operations can take: through the Micro-Instruction Translation Engine (MITE), through the Decode Stream Buffer (DSB), also called the Micro-operation Cache, or through the Loop Stream Detector (LSD). Each path has its own unique timing and power signatures, which lead to the side- and covert-channel attacks presented in this work. Especially, the switching between the different paths leads to observable timing or power differences which, as this work demonstrates, could be exploited by attackers. Because of the different paths, the switching, and way the components are shared in the frontend between hardware threads, two separate threads are able to be mutually influenced and timing or power can reveal activity on the other thread. The security threats are not limited to multi-threading, and this work further demonstrates new ways for leaking execution information about SGX enclaves or a new in-domain Spectre variant in single-thread setting. Finally, this work demonstrates a new method for fingerprinting the microcode patches of the processor by analyzing the behavior of different paths in the frontend. The findings of this work highlight the security threats associated with the processor frontend and the need for deployment of defenses for the modern processor frontend.

cs.CR↗

arXiv2021

Evaluation of Cache Attacks on Arm Processors and Secure Caches

Timing-based side and covert channels in processor caches continue to be a threat to modern computers. This work shows for the first time a systematic, large-scale analysis of Arm devices and the detailed results of attacks the processors are vulnerable to. Compared to x86, Arm uses different architectures, microarchitectural implementations, cache replacement policies, etc., which affects how attacks can be launched, and how security testing for the vulnerabilities should be done. To evaluate security, this paper presents security benchmarks specifically developed for testing Arm processors and their caches. The benchmarks are themselves evaluated with sensitivity tests, which examine how sensitive the benchmarks are to having a correct configuration in the testing phase. Further, to evaluate a large number of devices, this work leverages a novel approach of using a cloud-based Arm device testbed for architectural and security research on timing channels and runs the benchmarks on 34 different physical devices. In parallel, there has been much interest in secure caches to defend the various attacks. Consequently, this paper also investigates secure cache architectures using the proposed benchmarks. Especially, this paper implements and evaluates the secure PL and RF caches, showing the security of PL and RF caches, but also uncovers new weaknesses.

cs.CR↗

arXiv2019

A Benchmark Suite for Evaluating Caches' Vulnerability to Timing Attacks

Timing-based side or covert channels in processor caches continue to present a threat to computer systems, and they are the key to many of the recent Spectre and Meltdown attacks. Based on improvements to an existing three-step model for cache timing-based attacks, this work presents 88 Strong types of theoretical timing-based vulnerabilities in processor caches. To understand and evaluate all possible types of vulnerabilities in processor caches, this work further presents and implements a new benchmark suite which can be used to test to which types of cache timing-based attacks a given processor or cache design is vulnerable. In total, there are 1094 automatically-generated test programs which cover the 88 theoretical vulnerabilities. The benchmark suite generates the Cache Timing Vulnerability Score which can be used to evaluate how vulnerable a specific cache implementation is to different attacks. A smaller Cache Timing Vulnerability Score means the design is more secure, and the scores among different machines can be easily compared. Evaluation is conducted on commodity Intel and AMD processors and shows the differences in processor implementations can result in different types of attacks that they are vulnerable to. Beyond testing commodity processors, the benchmarks and the Cache Timing Vulnerability Score can be used to help designers of new secure processor caches evaluate their design's susceptibility to cache timing-based attacks.

cs.CR↗