Deep neural networks (DNNs) have been widely criticized for their overconfidence when dealing with out-of-distribution (OOD) samples, highlighting the critical need for effective OOD detection to ensure the safe deployment of DNNs in real-world settings. Existing post-hoc OOD detection methods primarily enhance the discriminative power of logit-based approaches by reshaping sample features, yet they often neglect critical information inherent in the features themselves. In this paper, we propose the Class-Aware Relative Feature-based method (CARef), which utilizes the error between a sample’s feature and its class-aware average feature as a discriminative criterion. To further refine this approach, we introduce the Class-Aware Decoupled Relative Feature-based method (CADRef), which decouples sample features based on the alignment of signs between the relative feature and corresponding model weights, enhancing the discriminative capabilities of CARef. Extensive experimental results across multiple datasets and models demonstrate that both proposed methods exhibit effectiveness and robustness in OOD detection compared to state-of-the-art methods. Specifically, our two methods outperform the best baseline by 2.82% and 3.27% in AUROC, with improvements of 4.03% and 6.32% in FPR95, respectively.

With the widespread adoption of open-source code language models (code LMs), intellectual property (IP) protection has become an increasingly critical concern.While current watermarking techniques have the potential to identify the code LM to protect its IP, they have limitations when facing the more practical and complex demand, i.e., offering the individual user-level tracing in the black-box setting.This work presents CLMTracing, a black-box code LM watermarking framework employing the rule-based watermarks and utility-preserving injection method for user-level model tracing.CLMTracing further incorporates a parameter selection algorithm sensitive to the robust watermark and adversarial training to enhance the robustness against watermark removal attacks.Comprehensive evaluations demonstrate CLM-Tracing is effective across multiple state-ofthe-art (SOTA) code LMs, showing significant harmless improvements compared to existing SOTA baselines and strong robustness against various removal attacks.

Sequence distance computation is a critical and fundamental task in many fields, such as bioinformatics, and time series analysis. Traditional functions for computing the distance between sequences are often based on dynamic programming to find a globally optimal alignment, which has quadratic complexity and is difficult to parallelize, thus limiting their application in large-scale datasets with long sequences. To solve this problem, various fields have designed some specialized models to approximate these distance functions inspired by deep representation learning, i.e., projecting the sequence into a geometric embedding space through an embedding function, so that the distance between sequences can be approximated by the distance in the high-dimensional embedding space, thereby reducing the quadratic complexity to linear. However, we note that even though the element types in sequence and distance functions are different across various fields, the core problem that needs to be solved remains the same. In this paper, we attempt to unify the sequence distance computation approximation from various fields and propose GnesDA. Specifically, we first unify the input representation of sequences in which the element type is the symbol and numeric values. We then encode the sequence using a convolutional block and a Transformer block sequentially, which can effectively capture local patterns and long dependencies respectively. Extensive experiments on four distance functions as well as four large-scale real-world datasets demonstrate that GnesDA achieves state-of-the-art in terms of both versatility and effectiveness. For the task of similarity retrieval, GnesDA can improve the edit distance, NW distance, DTW, and EDR by an average of 10.55%, 6.67%, 4.51%, and 12.00% on all metrics.

After large models (LMs) have gained widespread acceptance in code-related tasks, their superior generative capacity has greatly promoted the application of the code LM.Nevertheless, the security of the generated code has raised attention to its potential damage.Existing secure code generation methods have limited generalizability to unseen test cases and poor robustness against the attacked model, leading to safety failures in code generation.In this paper, we propose a generalizable and robust secure code generation method SecCoder by using in-context learning (ICL) and the safe demonstration.The dense retriever is also used to select the most helpful demonstration to maximize the improvement of the generated code's security.Experimental results show the superior generalizability of the proposed model Sec-Coder compared to the current secure code generation method, achieving a significant security improvement of an average of 7.20% on unseen test cases.The results also show the better robustness of SecCoder compared to the current attacked code LM, achieving a significant security improvement of an average of 7.74%.Our analysis indicates that SecCoder enhances the security of LMs in generating code, and it is more generalizable and robust.

Understanding the microarchitectural resource characteristics of datacenter jobs has become increasingly critical to guarantee the performance of jobs while improving resource utilization. Prior work studied the resource characteristics of datacenter jobs at the OS level, little reveals the deep and detailed characteristics at the microarchitecture level due to the lack of related open traces. In this paper, we provide a new open trace, AMTrace (Alibaba Microarchitecture Trace) 1, which is profiled from 8,577 high-end physical hosts from Alibaba’s datacenter by a hardware/software co-design monitoring method. AMTrace provides the microarchitectural metrics of 9.8 × 105 Linux containers with ”Per-Container-Per-Logic CPU” granularity. Different from existing open traces, AMTrace provides a new perspective to analyze the microarchitectural resource characteristics of datacenter jobs. Based on AMTrace, we first reveal the uneven resource usage of jobs among multiple logic CPUs. Then, we analyze the impact of resource contention of CPU and memory bandwidth on job performance. Finally, we analyze the job performance under different CPU provisioning modes from microarchitecture perspective. These analyses lead to constructive insights for datacenter resource management and optimization. Furthermore, we discuss possible research opportunities on AMTrace and we believe that AMTrace will inspire more exciting research on microarchitecture and resource management.

Java is a safe programming language by providing bytecode verification and enforcing memory protection. For instance, programmers cannot directly access the memory but have to use object references. Yet, the Java runtime provides an Unsafe API as a backdoor for the developers to access the low- level system code. Whereas the Unsafe API is designed to be used by the Java core library, a growing community of third-party libraries use it to achieve high performance. The Unsafe API is powerful, but dangerous, which leads to data corruption, resource leaks and difficult-to-diagnose JVM crash if used improperly. In this work, we study the Unsafe crash patterns and propose a memory checker to enforce memory safety, thus avoiding the JVM crash caused by the misuse of the Unsafe API at the bytecode level. We evaluate our technique on real crash cases from the openJDK bug system and real-world applications from AJDK. Our tool reduces the efforts from several days to a few minutes for the developers to diagnose the Unsafe related crashes. We also evaluate the runtime overhead of our tool on projects using intensive Unsafe operations, and the result shows that our tool causes a negligible perturbation to the execution of the applications.

The emergence of large-scale software deployments in the cloud has led to several challenges: (1) measuring software performance in the data center, and (2) optimizing software for resource management. This tutorial addresses the two challenges by bringing the knowledge of software performance monitoring in the data center to the world of applying performance analytics. It introduces data transformations for software performance metrics. The transformations enable effective applications of analytics. This tutorial starts with software performance in the small and ends with applying analytics to software performance in the large. In software performance in the small, it summarizes performance tools, data collection and manual analysis. Then it describes monitoring tools that are helpful in performance analysis in the large. The tutorial will guide the audience in applying analytics to performance data obtained by common tools. This tutorial describes how to select analytical methods and what precautions should be taken to get effective results.

Memory latency is a major factor in limiting CPU performance, and prefetching is a well-known method for hiding memory latency. Overly aggressive prefetching can waste scarce resources such as memory bandwidth and cache capacity, limiting or even hurting performance. It is therefore important to employ prefetching mechanisms that use these resources prudently, while still prefetching required data in a timely manner. In this work, we propose a new mechanism to determine at run-time the appropriate prefetching mechanism for the currently executing program, called Sandbox Prefetching. Sandbox Prefetching evaluates simple, aggressive offset prefetchers at run-time by adding the prefetch address to a Bloom filter, rather than actually fetching the data into the cache. Subsequent cache accesses are tested against the contents of the Bloom filter to see if the aggressive prefetcher under evaluation could have accurately prefetched the data, while simultaneously testing for the existence of prefetchable streams. Real prefetches are performed when the accuracy of evaluated prefetchers exceeds a threshold. This method combines the ideas of global pattern confirmation and immediate prefetching action to achieve high performance. Sandbox Prefetching improves performance across the tested workloads by 47.6% compared to not using any prefetching, and by 18.7% compared to the Feedback Directed Prefetching technique. Performance is also improved by 1.4% compared to the Access Map Pattern Matching Prefetcher, while incurring considerably less logic and storage overheads.

Software libraries provide leverage largely because they are used by many applications. As Parnas (1972, 1979), Lampson (1984) and others have noted, stable interfaces to libraries isolate the application from changes in the libraries. That is, as long as there is no change in a library's syntax or semantics, applications can use updated libraries simply by importing and linking the new version. However, libraries are indeed changed from time to time and the tedious job of adapting the application source to the library interface changes becomes a burden to multitudes of programmers. The paper introduces an approach and a toolset intended to reduce these costs. Specifically, in the authors' approach, a library maintainer annotates changed functions with rules that are used to generate tools that will update the applications that use the updated libraries. Thus, in exchange for a small added amount of work by the library maintainer, costs for each application maintainer can be reduced. They present the basic approach, describe the tools that support the approach, and discuss the strengths and limitation of the approach.