arXiv subjects

Shuai Zhao

Publications and source records attributed to Shuai Zhao.

At least 91 records · Page 5Linked to original sources

arXiv2025

MERE: Hardware-Software Co-Design for Masking Cache Miss Latency in Embedded Processors

Runahead execution is a technique to mask memory latency caused by irregular memory accesses. By pre-executing the application code during occurrences of long-latency operations and prefetching anticipated cache-missed data into the cache hierarchy, runahead effectively masks memory latency for subsequent cache misses and achieves high prefetching accuracy; however, this technique has been limited to superscalar out-of-order and superscalar in-order cores. For implementation in scalar in-order cores, the challenges of area-/energy-constraint and severe cache contention remain. Here, we build the first full-stack system featuring runahead, MERE, from SoC and a dedicated ISA to the OS and programming model. Through this deployment, we show that enabling runahead in scalar in-order cores is possible, with minimal area and power overheads, while still achieving high performance. By re-constructing the sequential runahead employing a hardware/software co-design approach, the system can be implemented on a mature processor and SoC. Building on this, an adaptive runahead mechanism is proposed to mitigate the severe cache contention in scalar in-order cores. Combining this, we provide a comprehensive solution for embedded processors managing irregular workloads. Our evaluation demonstrates that the proposed MERE attains 93.5% of a 2-wide out-of-order core's performance while constraining area and power overheads below 5%, with the adaptive runahead mechanism delivering an additional 20.1% performance gain through mitigating the severe cache contention issues.

cs.AR↗

arXiv2025

Exploiting the Uncertainty of the Longest Paths: Response Time Analysis for Probabilistic DAG Tasks

Parallel real-time systems (e.g., autonomous driving systems) often contain functionalities with complex dependencies and execution uncertainties, leading to significant timing variability which can be represented as a probabilistic distribution. However, existing timing analysis either produces a single conservative bound or suffers from severe scalability issues due to the exhaustive enumeration of every execution scenario. This causes significant difficulties in leveraging the probabilistic timing behaviours, resulting in sub-optimal design solutions. Modelling the system as a probabilistic directed acyclic graph (p-DAG), this paper presents a probabilistic response time analysis based on the longest paths of the p-DAG across all execution scenarios, enhancing the capability of the analysis by eliminating the need for enumeration. We first identify every longest path based on the structure of p-DAG and compute the probability of its occurrence. Then, the worst-case interfering workload is computed for each longest path, forming a complete probabilistic response time distribution with correctness guarantees. Experiments show that compared to the enumeration-based approach, the proposed analysis effectively scales to large p-DAGs with computation cost reduced by six orders of magnitude while maintaining a low deviation (1.04% on average and below 5% for most p-DAGs), empowering system design solutions with improved resource efficiency.

cs.DC↗

arXiv2025

Insights from Rights and Wrongs: A Large Language Model for Solving Assertion Failures in RTL Design

SystemVerilog Assertions (SVAs) are essential for verifying Register Transfer Level (RTL) designs, as they can be embedded into key functional paths to detect unintended behaviours. During simulation, assertion failures occur when the design's behaviour deviates from expectations. Solving these failures, i.e., identifying and fixing the issues causing the deviation, requires analysing complex logical and timing relationships between multiple signals. This process heavily relies on human expertise, and there is currently no automatic tool available to assist with it. Here, we present AssertSolver, an open-source Large Language Model (LLM) specifically designed for solving assertion failures. By leveraging synthetic training data and learning from error responses to challenging cases, AssertSolver achieves a bug-fixing pass@1 metric of 88.54% on our testbench, significantly outperforming OpenAI's o1-preview by up to 11.97%. We release our model and testbench for public access to encourage further research: https://github.com/SEU-ACAL/reproduce-AssertSolver-DAC-25.

cs.AR↗

arXiv2025

Unlocking a New Rust Programming Experience: Fast and Slow Thinking with LLMs to Conquer Undefined Behaviors

To provide flexibility and low-level interaction capabilities, the unsafe tag in Rust is essential in many projects, but undermines memory safety and introduces Undefined Behaviors (UBs) that reduce safety. Eliminating these UBs requires a deep understanding of Rust's safety rules and strong typing. Traditional methods require depth analysis of code, which is laborious and depends on knowledge design. The powerful semantic understanding capabilities of LLM offer new opportunities to solve this problem. Although existing large model debugging frameworks excel in semantic tasks, limited by fixed processes and lack adaptive and dynamic adjustment capabilities. Inspired by the dual process theory of decision-making (Fast and Slow Thinking), we present a LLM-based framework called RustBrain that automatically and flexibly minimizes UBs in Rust projects. Fast thinking extracts features to generate solutions, while slow thinking decomposes, verifies, and generalizes them abstractly. To apply verification and generalization results to solution generation, enabling dynamic adjustments and precise outputs, RustBrain integrates two thinking through a feedback mechanism. Experimental results on Miri dataset show a 94.3% pass rate and 80.4% execution rate, improving flexibility and Rust projects safety.

cs.SE↗

arXiv2025

Calculation of heavy meson light-cone distribution amplitudes from lattice QCD

We develop an approach for calculating heavy quark effective theory (HQET) light-cone distribution amplitudes (LCDAs) by employing a sequential effective theory methodology. The theoretical foundation of the framework is established, elucidating how the quasi distribution amplitudes (quasi DAs) with three scales can be utilized to compute HQET LCDAs. We provide theoretical support for this approach by demonstrating the rationale behind devising a hierarchical ordering for the three involved scales, discussing the factorization at each step, clarifying the underlying reason for obtaining HQET LCDAs in the final phase, and addressing potential theoretical challenges. The lattice QCD simulation aspect is explored in detail, and the computations of quasi DAs are presented. We employ three fitting strategies to handle contributions from excited states and extract the bare matrix elements. For renormalization purposes, we apply hybrid renormalization schemes at short and long distance separations. To mitigate long-distance perturbations, we perform an extrapolation in $λ= z\cdot P^z$ and assess the stability against various parameters. After two-step matching, our results for HQET LCDAs are found in agreement with existing model parametrizations. The potential phenomenological implications of the results are discussed, shedding light on how these findings could impact our understanding of the strong interaction dynamics and physics beyond the standard model. It should be noted, however, that systematic uncertainties have not been accounted for yet.

hep-lat↗

arXiv2025

Baichuan-Omni-1.5 Technical Report

We introduce Baichuan-Omni-1.5, an omni-modal model that not only has omni-modal understanding capabilities but also provides end-to-end audio generation capabilities. To achieve fluent and high-quality interaction across modalities without compromising the capabilities of any modality, we prioritized optimizing three key aspects. First, we establish a comprehensive data cleaning and synthesis pipeline for multimodal data, obtaining about 500B high-quality data (text, audio, and vision). Second, an audio-tokenizer (Baichuan-Audio-Tokenizer) has been designed to capture both semantic and acoustic information from audio, enabling seamless integration and enhanced compatibility with MLLM. Lastly, we designed a multi-stage training strategy that progressively integrates multimodal alignment and multitask fine-tuning, ensuring effective synergy across all modalities. Baichuan-Omni-1.5 leads contemporary models (including GPT4o-mini and MiniCPM-o 2.6) in terms of comprehensive omni-modal capabilities. Notably, it achieves results comparable to leading models such as Qwen2-VL-72B across various multimodal medical benchmarks.

cs.CL↗

arXiv2025

Enhancing Multimodal Entity Linking with Jaccard Distance-based Conditional Contrastive Learning and Contextual Visual Augmentation

Previous research on multimodal entity linking (MEL) has primarily employed contrastive learning as the primary objective. However, using the rest of the batch as negative samples without careful consideration, these studies risk leveraging easy features and potentially overlook essential details that make entities unique. In this work, we propose JD-CCL (Jaccard Distance-based Conditional Contrastive Learning), a novel approach designed to enhance the ability to match multimodal entity linking models. JD-CCL leverages meta-information to select negative samples with similar attributes, making the linking task more challenging and robust. Additionally, to address the limitations caused by the variations within the visual modality among mentions and entities, we introduce a novel method, CVaCPT (Contextual Visual-aid Controllable Patch Transform). It enhances visual representations by incorporating multi-view synthetic images and contextual textual representations to scale and shift patch representations. Experimental results on benchmark MEL datasets demonstrate the strong effectiveness of our approach.

cs.CV↗

arXiv2025

Linear relation between short range correlation and EMC effect of gluons in nuclei

We explore the EMC effect of gluons through heavy quark production in deep inelastic scattering (DIS) and the nucleon-nucleon short range correlation (SRC) from sub-threshold photoproduction of the $J/ψ$. Applying an effective field theory analysis, we scrutinize the gluon parton distribution functions in nuclear environment and propose a linear relation between the slope of reduced cross section ratio, which reflects the magnitude of gluon EMC effect, and the $J/ψ$ production cross section at the photon energy $E_γ= 7.0\,\textrm{GeV}$ in the nucleus rest frame, a characteristic representative of SRC. This finding is supported by a numerical calculation using the available phenomenological results for gluon nuclear parton distribution functions (nPDFs). Examining this linear relation by the forthcoming experimental data on electron-ion collider allows to decipher nuclear effects of gluon distributions and enhance our comprehension of the substructure in bound nuclei.

hep-ph↗

arXiv2025

A Survey of Recent Backdoor Attacks and Defenses in Large Language Models

Large Language Models (LLMs), which bridge the gap between human language understanding and complex problem-solving, achieve state-of-the-art performance on several NLP tasks, particularly in few-shot and zero-shot settings. Despite the demonstrable efficacy of LLMs, due to constraints on computational resources, users have to engage with open-source language models or outsource the entire training process to third-party platforms. However, research has demonstrated that language models are susceptible to potential security vulnerabilities, particularly in backdoor attacks. Backdoor attacks are designed to introduce targeted vulnerabilities into language models by poisoning training samples or model weights, allowing attackers to manipulate model responses through malicious triggers. While existing surveys on backdoor attacks provide a comprehensive overview, they lack an in-depth examination of backdoor attacks specifically targeting LLMs. To bridge this gap and grasp the latest trends in the field, this paper presents a novel perspective on backdoor attacks for LLMs by focusing on fine-tuning methods. Specifically, we systematically classify backdoor attacks into three categories: full-parameter fine-tuning, parameter-efficient fine-tuning, and no fine-tuning Based on insights from a substantial review, we also discuss crucial issues for future research on backdoor attacks, such as further exploring attack algorithms that do not require fine-tuning, or developing more covert attack algorithms.

cs.CR↗

arXiv2024

CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model

Pre-trained vision-language models~(VLMs) are the de-facto foundation models for various downstream tasks. However, scene text recognition methods still prefer backbones pre-trained on a single modality, namely, the visual modality, despite the potential of VLMs to serve as powerful scene text readers. For example, CLIP can robustly identify regular (horizontal) and irregular (rotated, curved, blurred, or occluded) text in images. With such merits, we transform CLIP into a scene text reader and introduce CLIP4STR, a simple yet effective STR method built upon image and text encoders of CLIP. It has two encoder-decoder branches: a visual branch and a cross-modal branch. The visual branch provides an initial prediction based on the visual feature, and the cross-modal branch refines this prediction by addressing the discrepancy between the visual feature and text semantics. To fully leverage the capabilities of both branches, we design a dual predict-and-refine decoding scheme for inference. We scale CLIP4STR in terms of the model size, pre-training data, and training data, achieving state-of-the-art performance on 13 STR benchmarks. Additionally, a comprehensive empirical study is provided to enhance the understanding of the adaptation of CLIP to STR. Our method establishes a simple yet strong baseline for future STR research with VLMs.

cs.CV↗

arXiv2024

Creating true muonium via charmonium radiative decay

True muonium, the bound state of a muon and an antimuon ($μ^+μ^-$), has long been theoretically predicted but remains experimentally elusive. We investigate the production of true para-muonium in the radiative decay of $J/ψ$ meson,and analyze the prospects for detecting true muonium in current and future high-energy $e^+e^-$ experiments, particularly focusing on the BESIII experiment and the proposed Super Tau-Charm Facility. Although the events are rare at the super tau-charm facility, the detection of true para-muonium via $J/ψ$ radiative decays could become feasible at its future updates.

hep-ph↗

arXiv2024

Mass renormalization group of heavy meson light-cone distribution amplitude in QCD

The heavy meson light-cone distribution amplitude (LCDA), as defined in full QCD, plays a key role in the collinear factorization for exclusive heavy meson production and in lattice computations of the LCDA within heavy-quark effective theory (HQET). In addition to its dependence on the renormalization scale, the QCD LCDA also evolves with the heavy quark mass. We derive a partial differential equation to characterize the mass evolution of the heavy meson QCD LCDA, examining the heavy quark mass dependence through its solution. Our results link the internal structure of heavy mesons across different quark masses, offering significant implications for lattice calculations and enabling the extrapolation of results from lower to higher quark masses.

hep-ph↗

arXiv2024

Cluster-based Graph Collaborative Filtering

Graph Convolution Networks (GCNs) have significantly succeeded in learning user and item representations for recommendation systems. The core of their efficacy is the ability to explicitly exploit the collaborative signals from both the first- and high-order neighboring nodes. However, most existing GCN-based methods overlook the multiple interests of users while performing high-order graph convolution. Thus, the noisy information from unreliable neighbor nodes (e.g., users with dissimilar interests) negatively impacts the representation learning of the target node. Additionally, conducting graph convolution operations without differentiating high-order neighbors suffers the over-smoothing issue when stacking more layers, resulting in performance degradation. In this paper, we aim to capture more valuable information from high-order neighboring nodes while avoiding noise for better representation learning of the target node. To achieve this goal, we propose a novel GCN-based recommendation model, termed Cluster-based Graph Collaborative Filtering (ClusterGCF). This model performs high-order graph convolution on cluster-specific graphs, which are constructed by capturing the multiple interests of users and identifying the common interests among them. Specifically, we design an unsupervised and optimizable soft node clustering approach to classify user and item nodes into multiple clusters. Based on the soft node clustering results and the topology of the user-item interaction graph, we assign the nodes with probabilities for different clusters to construct the cluster-specific graphs. To evaluate the effectiveness of ClusterGCF, we conducted extensive experiments on four publicly available datasets. Experimental results demonstrate that our model can significantly improve recommendation performance.

cs.IR↗

arXiv2024

Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning

In-context learning, a paradigm bridging the gap between pre-training and fine-tuning, has demonstrated high efficacy in several NLP tasks, especially in few-shot settings. Despite being widely applied, in-context learning is vulnerable to malicious attacks. In this work, we raise security concerns regarding this paradigm. Our studies demonstrate that an attacker can manipulate the behavior of large language models by poisoning the demonstration context, without the need for fine-tuning the model. Specifically, we design a new backdoor attack method, named ICLAttack, to target large language models based on in-context learning. Our method encompasses two types of attacks: poisoning demonstration examples and poisoning demonstration prompts, which can make models behave in alignment with predefined intentions. ICLAttack does not require additional fine-tuning to implant a backdoor, thus preserving the model's generality. Furthermore, the poisoned examples are correctly labeled, enhancing the natural stealth of our attack method. Extensive experimental results across several language models, ranging in size from 1.3B to 180B parameters, demonstrate the effectiveness of our attack method, exemplified by a high average attack success rate of 95.0% across the three datasets on OPT models.

cs.CL↗

arXiv2024

Hardware/Algorithm Co-design for Real-Time I/O Control with Improved Timing Accuracy and Robustness

In safety-critical systems, timing accuracy is the key to achieving precise I/O control. To meet such strict timing requirements, dedicated hardware assistance has recently been investigated and developed. However, these solutions are often fragile, due to unforeseen timing defects. In this paper, we propose a robust and timing-accurate I/O co-processor, which manages I/O tasks using Execution Time Servers (ETSs) and a two-level scheduler. The ETSs limit the impact of timing defects between tasks, and the scheduler prioritises ETSs based on their importance, offering a robust and configurable scheduling infrastructure. Based on the hardware design, we present an ETS-based timing-accurate I/O schedule, with the ETS parameters configured to further enhance robustness against timing defects. Experiments show the proposed I/O control method outperforms the state-of-the-art method in terms of timing accuracy and robustness without introducing significant overhead.

cs.AR↗

arXiv2024

FRAP: A Flexible Resource Accessing Protocol for Multiprocessor Real-Time Systems

Fully-partitioned fixed-priority scheduling (FP-FPS) multiprocessor systems are widely found in real-time applications, where spin-based protocols are often deployed to manage the mutually exclusive access of shared resources. Unfortunately, existing approaches either enforce rigid spin priority rules for resource accessing or carry significant pessimism in the schedulability analysis, imposing substantial blocking time regardless of task execution urgency or resource over-provisioning. This paper proposes FRAP, a spin-based flexible resource accessing protocol for FP-FPS systems. A task under FRAP can spin at any priority within a range for accessing a resource, allowing flexible and fine-grained resource control with predictable worst-case behaviour. Under flexible spinning, we demonstrate that the existing analysis techniques can lead to incorrect timing bounds and present a novel MCMF (minimum cost maximum flow)-based blocking analysis, providing predictability guarantee for FRAP. A spin priority assignment is reported that fully exploits flexible spinning to reduce the blocking time of tasks with high urgency, enhancing the performance of FRAP. Experimental results show that FRAP outperforms the existing spin-based protocols in schedulability by 15.20%-32.73% on average, up to 65.85%.

cs.OS↗

arXiv2024

Decoder Pre-Training with only Text for Scene Text Recognition

Scene text recognition (STR) pre-training methods have achieved remarkable progress, primarily relying on synthetic datasets. However, the domain gap between synthetic and real images poses a challenge in acquiring feature representations that align well with images on real scenes, thereby limiting the performance of these methods. We note that vision-language models like CLIP, pre-trained on extensive real image-text pairs, effectively align images and text in a unified embedding space, suggesting the potential to derive the representations of real images from text alone. Building upon this premise, we introduce a novel method named Decoder Pre-training with only text for STR (DPTR). DPTR treats text embeddings produced by the CLIP text encoder as pseudo visual embeddings and uses them to pre-train the decoder. An Offline Randomized Perturbation (ORP) strategy is introduced. It enriches the diversity of text embeddings by incorporating natural image embeddings extracted from the CLIP image encoder, effectively directing the decoder to acquire the potential representations of real images. In addition, we introduce a Feature Merge Unit (FMU) that guides the extracted visual embeddings focusing on the character foreground within the text image, thereby enabling the pre-trained decoder to work more efficiently and accurately. Extensive experiments across various STR decoders and language recognition tasks underscore the broad applicability and remarkable performance of DPTR, providing a novel insight for STR pre-training. Code is available at https://github.com/Topdu/OpenOCR

cs.CV↗

arXiv2024

Slimmable Networks for Contrastive Self-supervised Learning

Self-supervised learning makes significant progress in pre-training large models, but struggles with small models. Mainstream solutions to this problem rely mainly on knowledge distillation, which involves a two-stage procedure: first training a large teacher model and then distilling it to improve the generalization ability of smaller ones. In this work, we introduce another one-stage solution to obtain pre-trained small models without the need for extra teachers, namely, slimmable networks for contrastive self-supervised learning (SlimCLR). A slimmable network consists of a full network and several weight-sharing sub-networks, which can be pre-trained once to obtain various networks, including small ones with low computation costs. However, interference between weight-sharing networks leads to severe performance degradation in self-supervised cases, as evidenced by gradient magnitude imbalance and gradient direction divergence. The former indicates that a small proportion of parameters produce dominant gradients during backpropagation, while the main parameters may not be fully optimized. The latter shows that the gradient direction is disordered, and the optimization process is unstable. To address these issues, we introduce three techniques to make the main parameters produce dominant gradients and sub-networks have consistent outputs. These techniques include slow start training of sub-networks, online distillation, and loss re-weighting according to model sizes. Furthermore, theoretical results are presented to demonstrate that a single slimmable linear layer is sub-optimal during linear evaluation. Thus a switchable linear probe layer is applied during linear evaluation. We instantiate SlimCLR with typical contrastive learning frameworks and achieve better performance than previous arts with fewer parameters and FLOPs. The code is at https://github.com/mzhaoshuai/SlimCLR.

cs.CV↗