Shuai Wang

Zhenlan Ji, Pingchuan Ma, Zongjie Li et al.

While code generation has been widely used in various software development scenarios, the quality of the generated code is not guaranteed. This has been a particular concern in the era of large language models (LLMs)- based code generation, where LLMs, deemed a complex and powerful black-box model, is instructed by a high-level natural language specification, namely a prompt, to generate code. Nevertheless, effectively evaluating and explaining the code generation capability of LLMs is inherently challenging, given the complexity of LLMs and the lack of transparency. Inspired by the recent progress in causality analysis and its application in software engineering, this paper launches a causality analysis-based approach to systematically analyze the causal relations between the LLM input prompts and the generated code. To handle various technical challenges in this study, we first propose a novel causal graph-based representation of the prompt and the generated code, which is established over the fine-grained, human-understandable concepts in the input prompts. The formed causal graph is then used to identify the causal relations between the prompt and the derived code. We illustrate the insights that our framework can provide by studying over 3 popular LLMs with over 12 prompt adjustment strategies. The results of these studies illustrate the potential of our technique to provide insights into LLM effectiveness, and aid end-users in understanding predictions. Additionally, we demonstrate that our approach provides actionable insights to improve the quality of the LLM-generated code by properly calibrating the prompt.

Zongjie Li, Daoyuan Wu, Shuai Wang et al.

Large code models (LCMs), pre-trained on vast code corpora, have demonstrated remarkable performance across a wide array of code-related tasks. Supervised fine-tuning (SFT) plays a vital role in aligning these models with specific requirements and enhancing their performance in particular domains. However, synthesizing high-quality SFT datasets poses a significant challenge due to the uneven quality of datasets and the scarcity of domain-specific datasets. Inspired by APIs as high-level abstractions of code that encapsulate rich semantic information in a concise structure, we propose DataScope, an API-guided dataset synthesis framework designed to enhance the SFT process for LCMs in both general and domain-specific scenarios. DataScope comprises two main components: Dsel and Dgen. On one hand, Dsel employs API coverage as a core metric, enabling efficient dataset synthesis in general scenarios by selecting subsets of existing (uneven-quality) datasets with higher API coverage. On the other hand, Dgen recasts domain dataset synthesis as a process of using API-specified high-level functionality and deliberately-constituted code skeletons to synthesize concrete code. Extensive experiments demonstrate DataScope's effectiveness, with models fine-tuned on its synthesized datasets outperforming those tuned on unoptimized datasets five times larger. Furthermore, a series of analyses on model internals, relevant hyperparameters, and case studies provide additional evidence for the efficacy of our proposed methods. These findings underscore the significance of dataset quality in SFT and advance the field of LCMs by providing an efficient, cost-effective framework for constructing high-quality datasets. This contribution enhances performance across both general and domain-specific scenarios, paving the way for more powerful and tailored LCMs.

Yanzuo Chen, Zhibo Liu, Yuanyuan Yuan et al.

Bit-flip attacks (BFAs) can manipulate deep neural networks (DNNs). For high-level DNN models running on deep learning (DL) frameworks like PyTorch, extensive BFAs have been used to flip bits in model weights and shown effective. Defenses have also been proposed to guard model weights. However, DNNs are increasingly compiled into DNN executables by DL compilers to leverage hardware primitives. These executables manifest distinct computation paradigms; existing research fails to accurately capture and expose the BFA surfaces on DNN executables. To this end, we launch the first systematic study of BFAs on DNN executables. Prior BFAs are limited to attacking model weights and assume a strong whitebox attacker with full knowledge of victim model weights, which is unrealistic as weights are often confidential. In contrast, we find that BFAs on DNN executables can achieve high effectiveness by exploiting the model structure (usually stored in the executable code), which only requires knowing the (often public) model structure. Importantly, such structure-based BFAs are pervasive, transferable, and more severe in DNN executables. They also slip past existing defenses. To demonstrate the new attack surfaces, we assume a weak and more realistic attacker with no knowledge of victim model weights. We design an automated tool to identify vulnerable bits in victim executables with high confidence (70% vs. baseline 2%). We show on DDR4 DRAM that only 1.4 flips on average are needed to fully downgrade the accuracy of victim models, including quantized ones which could require 23x more flips previously, to random guesses. We comprehensively evaluate 16 DNN executables, covering large-scale models trained on commonly-used datasets compiled by the two most popular DL compilers. Our finding calls for incorporating security mechanisms in future DNN compilation toolchains.

Xun Lin, Shuai Wang, Rizhao Cai et al.

Face Anti-Spoofing (FAS) is crucial for securing face recognition systems against presentation attacks. With advancements in sensor manufacture and multi-modal learning techniques, many multi-modal FAS approaches have emerged. However, they face challenges in generalizing to unseen attacks and deployment conditions. These challenges arise from (1) modality unreliability, where some modality sensors like depth and infrared undergo significant domain shifts in varying environments, leading to the spread of unreliable information during cross-modal feature fusion, and (2) modality imbalance, where training overly relies on a dominant modality hinders the convergence of others, reducing effectiveness against attack types that are indistinguishable sorely using the dominant modality. To address modality unreliability, we propose the Uncertainty-Guided Cross-Adapter (U-Adapter) to recognize unreliably detected regions within each modality and suppress the impact of unreliable regions on other modalities. For modality imbalance, we propose a Rebalanced Modality Gradient Modulation (ReGrad) strategy to rebalance the convergence speed of all modalities by adaptively adjusting their gradients. Besides, we provide the first large-scale benchmark for evaluating multi-modal FAS performance under domain generalization scenarios. Extensive experiments demonstrate that our method outperforms state-of-the-art methods. Source code and protocols will be released on https://github.com/OMGGGGG/mmdg.

Shuai Wang, Liang Ding, Li Shen et al.

Advancing automated programming necessitates robust and comprehensive code generation benchmarks, yet current evaluation frameworks largely neglect object-oriented programming (OOP) in favor of functional programming (FP), e.g., HumanEval and MBPP. To address this, our study introduces a pioneering OOP-focused benchmark, featuring 431 Python programs that encompass essential OOP concepts and features like classes and encapsulation methods. We propose a novel evaluation metric, pass@o, tailored for OOP, enhancing traditional pass@k measures. Our evaluation of 23 leading large language models (LLMs), including both general and code-specialized models, reveals three key insights: 1) pass@o offers a more relevant and comprehensive assessment for OOP code generation; 2) Despite excelling in FP, code-specialized LLMs like WizardCoder lag in OOP compared to models like ChatGPT; 3) The poor performance of all advanced LLMs on our OOP benchmark highlights a critical need for improvements in this field. Our benchmark and scripts are publicly released at: https://github.com/alphadl/OOP-eval.

Tiannan Wang, Meiling Tao, Ruoyu Fang et al.

In this work, we introduce the task of life-long personalization of large language models. While recent mainstream efforts in the LLM community mainly focus on scaling data and compute for improved capabilities of LLMs, we argue that it is also very important to enable LLM systems, or language agents, to continuously adapt to the diverse and ever-changing profiles of every distinct user and provide up-to-date personalized assistance. We provide a clear task formulation and introduce a simple, general, effective, and scalable framework for life-long personalization of LLM systems and language agents. To facilitate future research on LLM personalization, we also introduce methods to synthesize realistic benchmarks and robust evaluation metrics. We will release all codes and data for building and benchmarking life-long personalized LLM systems.

Fei Wang, Chao Shang, Sarthak Jain et al.

User alignment is crucial for adapting general-purpose language models (LMs) to downstream tasks, but human annotations are often not available for all types of instructions, especially those with customized constraints. We observe that user instructions typically contain constraints. While assessing response quality in terms of the whole instruction is often costly, efficiently evaluating the satisfaction rate of constraints is feasible. We investigate common constraints in NLP tasks, categorize them into three classes based on the types of their arguments, and propose a unified framework, ACT (Aligning to ConsTraints), to automatically produce supervision signals for user alignment with constraints. Specifically, ACT uses constraint verifiers, which are typically easy to implement in practice, to compute constraint satisfaction rate (CSR) of each response. It samples multiple responses for each prompt and collect preference labels based on their CSR automatically. Subsequently, ACT adapts the LM to the target task through a ranking-based learning process. Experiments on fine-grained entity typing, abstractive summarization, and temporal question answering show that ACT is able to enhance LMs' capability to adhere to different classes of constraints, thereby improving task performance. Further experiments show that the constraint-following capabilities are transferable.

Zongjie Li, Shuai Wang

This position paper proposes a fundamental shift in designing code generation models: treating reasoning depth as a controllable resource. Rather than being an incidental byproduct of prompting, we argue that the trade-off between rapid, direct answers ("fast thinking") and elaborate, chain-of-thought deliberation ("slow thinking") must be explicitly managed. We contend that optimizing reasoning budgets across the entire model lifecycle - from synthetic data creation and benchmarking to real-world deploymen - can unlock superior trade-offs among accuracy, latency, and cost. This paper outlines how adaptive control over reasoning can enrich supervision signals, motivate new multi-dimensional benchmarks, and inform cost-aware, security-conscious deployment policies. By viewing fast and slow thinking as complementary modes to be scheduled, we envision coding agents that think deep when necessary and act fast when possible.

Wenxuan Wang, Yongjiang Wu, Junyuan Zhang et al.

The rapid growth of audio-centric platforms and applications such as WhatsApp and Twitter has transformed the way people communicate and share audio content in modern society. However, these platforms are increasingly misused to disseminate harmful audio content, such as hate speech, deceptive advertisements, and explicit material, which can have significant negative consequences (e.g., detrimental effects on mental health). In response, researchers and practitioners have been actively developing and deploying audio content moderation tools to tackle this issue. Despite these efforts, malicious actors can bypass moderation systems by making subtle alterations to audio content, such as modifying pitch or inserting noise. Moreover, the effectiveness of modern audio moderation tools against such adversarial inputs remains insufficiently studied. To address these challenges, we propose MTAM, a Metamorphic Testing framework for Audio content Moderation software. Specifically, we conduct a pilot study on 2000 audio clips and define 14 metamorphic relations across two perturbation categories: Audio Features-Based and Heuristic perturbations. MTAM applies these metamorphic relations to toxic audio content to generate test cases that remain harmful while being more likely to evade detection. In our evaluation, we employ MTAM to test five commercial textual content moderation software and an academic model against three kinds of toxic content. The results show that MTAM achieves up to 38.6%, 18.3%, 35.1%, 16.7%, and 51.1% error finding rates (EFR) when testing commercial moderation software provided by Gladia, Assembly AI, Baidu, Nextdata, and Tencent, respectively, and it obtains up to 45.7% EFR when testing the state-of-the-art algorithms from the academy.

Guanyuan Pan, Tiansheng Zhou, Jianxiang Zhao et al.

Circuit link prediction, which identifies missing component connections from incomplete netlists, is crucial in analog circuit design automation. However, existing methods face three main challenges: 1) Insufficient use of topological patterns in circuit graphs reduces prediction accuracy; 2) Data scarcity due to the complexity of annotations hinders model generalization; 3) Limited adaptability to various netlist formats restricts model flexibility. We propose Graph Neural Networks Based Analog Circuit Link Prediction (GNN-ACLP), a graph neural networks (GNNs) based method featuring three innovations to tackle these challenges. First, we introduce the SEAL (learning from Subgraphs, Embeddings, and Attributes for Link prediction) framework and achieve port-level accuracy in circuit link prediction. Second, we propose Netlist Babel Fish, a netlist format conversion tool that leverages retrieval-augmented generation (RAG) with a large language model (LLM) to enhance the compatibility of netlist formats. Finally, we build a comprehensive dataset, SpiceNetlist, comprising 775 annotated circuits of 7 different types across 10 component classes. Experiments demonstrate accuracy improvements of 16.08% on SpiceNetlist, 11.38% on Image2Net, and 16.01% on Masala-CHAI compared to the baseline in intra-dataset evaluation, while maintaining accuracy from 92.05% to 99.07% in cross-dataset evaluation, demonstrating robust feature transfer capabilities. However, its linear computational complexity makes processing large-scale netlists challenging and requires future addressing.

Wangchunshu Zhou, Yixin Ou, Shengwei Ding et al.

The AI community has been exploring a pathway to artificial general intelligence (AGI) by developing "language agents", which are complex large language models (LLMs) pipelines involving both prompting techniques and tool usage methods. While language agents have demonstrated impressive capabilities for many real-world tasks, a fundamental limitation of current language agents research is that they are model-centric, or engineering-centric. That's to say, the progress on prompts, tools, and pipelines of language agents requires substantial manual engineering efforts from human experts rather than automatically learning from data. We believe the transition from model-centric, or engineering-centric, to data-centric, i.e., the ability of language agents to autonomously learn and evolve in environments, is the key for them to possibly achieve AGI. In this work, we introduce agent symbolic learning, a systematic framework that enables language agents to optimize themselves on their own in a data-centric way using symbolic optimizers. Specifically, we consider agents as symbolic networks where learnable weights are defined by prompts, tools, and the way they are stacked together. Agent symbolic learning is designed to optimize the symbolic network within language agents by mimicking two fundamental algorithms in connectionist learning: back-propagation and gradient descent. Instead of dealing with numeric weights, agent symbolic learning works with natural language simulacrums of weights, loss, and gradients. We conduct proof-of-concept experiments on both standard benchmarks and complex real-world tasks and show that agent symbolic learning enables language agents to update themselves after being created and deployed in the wild, resulting in "self-evolving agents".

Jiazhao Wang, Wenchao Jiang, Ruofeng Liu et al.

A physical-layer modulator is a vital component for an IoT gateway to map the symbols to signals. However, due to the soldered hardware chipsets on the gateway's motherboards or the diverse toolkits on different platforms for the software radio, the existing solutions either have limited extensibility or are platform-specific. Such limitation is hard to ignore when modulation schemes and hardware platforms have become extremely diverse. This paper presents a new paradigm of using neural networks as an abstraction layer for physical layer modulators in IoT gateway devices, referred to as NN-defined modulators. Our approach addresses the challenges of extensibility and portability for multiple technologies on various hardware platforms. The proposed NN-defined modulator uses a model-driven methodology rooted in solid mathematical foundations while having native support for hardware acceleration and portability to heterogeneous platforms. We conduct the evaluation of NN-defined modulators on different platforms, including Nvidia Jetson Nano and Raspberry Pi. Evaluations demonstrate that our NN-defined modulator effectively operates as conventional modulators and provides significant efficiency gains (up to $4.7\times$ on Nvidia Jetson Nano and $1.1\times$ on Raspberry Pi), indicating high portability. Furthermore, we show the real-world applications using our NN-defined modulators to generate ZigBee and WiFi packets, which are compliant with commodity TI CC2650 (ZigBee) and Intel AX201 (WiFi NIC), respectively.

Meng Ge, Yizhou Peng, Yidi Jiang et al.

This paper summarizes our team's efforts in both tracks of the ICMC-ASR Challenge for in-car multi-channel automatic speech recognition. Our submitted systems for ICMC-ASR Challenge include the multi-channel front-end enhancement and diarization, training data augmentation, speech recognition modeling with multi-channel branches. Tested on the offical Eval1 and Eval2 set, our best system achieves a relative 34.3% improvement in CER and 56.5% improvement in cpCER, compared to the offical baseline system.

Yuanyuan Yuan, Qi Pang, Shuai Wang

The prosperous development of cloud computing and machine learning as a service has led to the widespread use of media software to process confidential media data. This paper explores an adversary's ability to launch side channel analyses (SCA) against media software to reconstruct confidential media inputs. Recent advances in representation learning and perceptual learning inspired us to consider the reconstruction of media inputs from side channel traces as a cross-modality manifold learning task that can be addressed in a unified manner with an autoencoder framework trained to learn the mapping between media inputs and side channel observations. We further enhance the autoencoder with attention to localize the program points that make the primary contribution to SCA, thus automatically pinpointing information-leakage points in media software. We also propose a novel and highly effective defensive technique called perception blinding that can perturb media inputs with perception masks and mitigate manifold learning-based SCA. Our evaluation exploits three popular media software to reconstruct inputs in image, audio, and text formats. We analyze three common side channels - cache bank, cache line, and page tables - and userspace-only cache set accesses logged by standard Prime+Probe. Our framework successfully reconstructs high-quality confidential inputs from the assessed media software and automatically pinpoint their vulnerable program points, many of which are unknown to the public. We further show that perception blinding can mitigate manifold learning-based SCA with negligible extra cost.

Pingchuan Ma, Shuai Wang

Text-to-SQL is a task to generate SQL queries from human utterances. However, due to the variation of natural language, two semantically equivalent utterances may appear differently in the lexical level. Likewise, user preferences (e.g., the choice of normal forms) can lead to dramatic changes in table structures when expressing conceptually identical schemas. Envisioning the general difficulty for text-to-SQL models to preserve prediction consistency against linguistic and schema variations, we propose MT-Teql, a Metamorphic Testing-based framework for systematically evaluating and augmenting the consistency of TExt-to-SQL models. Inspired by the principles of software metamorphic testing, MT-Teql delivers a model-agnostic framework which implements a comprehensive set of metamorphic relations (MRs) to conduct semantics-preserving transformations toward utterances and schemas. Model Inconsistency can be exposed when the original and transformed inputs induce different SQL queries. In addition, we leverage the transformed inputs to retrain models for further model robustness boost. Our experiments show that our framework exposes thousands of prediction errors from SOTA models and enriches existing datasets by order of magnitude, eliminating over 40% inconsistency errors without compromising standard accuracy.

Sahisnu Mazumder, Bing Liu, Shuai Wang et al.

Traditional approaches to building natural language (NL) interfaces typically use a semantic parser to parse the user command and convert it to a logical form, which is then translated to an executable action in an application. However, it is still challenging for a semantic parser to correctly parse natural language. For a different domain, the parser may need to be retrained or tuned, and a new translator also needs to be written to convert the logical forms to executable actions. In this work, we propose a novel and application independent approach to building NL interfaces that does not need a semantic parser or a translator. It is based on natural language to natural language matching and learning, where the representation of each action and each user command are both in natural language. To perform a user intended action, the system only needs to match the user command with the correct action representation, and then execute the corresponding action. The system also interactively learns new (paraphrased) commands for actions to expand the action representations over time. Our experimental results show the effectiveness of the proposed approach.

Shuai Wang, Yuyan Bao, Xiao Liu et al.

Cache-based side channels enable a dedicated attacker to reveal program secrets by measuring the cache access patterns. Practical attacks have been shown against real-world crypto algorithm implementations such as RSA, AES, and ElGamal. By far, identifying information leaks due to cache-based side channels, either in a static or dynamic manner, remains a challenge: the existing approaches fail to offer high precision, full coverage, and good scalability simultaneously, thus impeding their practical use in real-world scenarios. In this paper, we propose a novel static analysis method on binaries to detect cache-based side channels. We use abstract interpretation to reason on program states with respect to abstract values at each program point. To make such abstract interpretation scalable to real-world cryptosystems while offering high precision and full coverage, we propose a novel abstract domain called the Secret-Augmented Symbolic domain (SAS). SAS tracks program secrets and dependencies on them for precision, while it tracks only coarse-grained public information for scalability. We have implemented the proposed technique into a practical tool named CacheS and evaluated it on the implementations of widely-used cryptographic algorithms in real-world crypto libraries, including Libgcrypt, OpenSSL, and mbedTLS. CacheS successfully confirmed a total of 154 information leaks reported by previous research and 54 leaks that were previously unknown. We have reported our findings to the developers. And they confirmed that many of those unknown information leaks do lead to potential side channels.

Le Guan, Jun Xu, Shuai Wang et al.

Nowadays, auto insurance companies set personalized insurance rate based on data gathered directly from their customers' cars. In this paper, we show such a personalized insurance mechanism -- wildly adopted by many auto insurance companies -- is vulnerable to exploit. In particular, we demonstrate that an adversary can leverage off-the-shelf hardware to manipulate the data to the device that collects drivers' habits for insurance rate customization and obtain a fraudulent insurance discount. In response to this type of attack, we also propose a defense mechanism that escalates the protection for insurers' data collection. The main idea of this mechanism is to augment the insurer's data collection device with the ability to gather unforgeable data acquired from the physical world, and then leverage these data to identify manipulated data points. Our defense mechanism leveraged a statistical model built on unmanipulated data and is robust to manipulation methods that are not foreseen previously. We have implemented this defense mechanism as a proof-of-concept prototype and tested its effectiveness in the real world. Our evaluation shows that our defense mechanism exhibits a false positive rate of 0.032 and a false negative rate of 0.013.

Pei Wang, Shuai Wang, Jiang Ming et al.

Program obfuscation is an important software protection technique that prevents attackers from revealing the programming logic and design of the software. We introduce translingual obfuscation, a new software obfuscation scheme which makes programs obscure by "misusing" the unique features of certain programming languages. Translingual obfuscation translates part of a program from its original language to another language which has a different programming paradigm and execution model, thus increasing program complexity and impeding reverse engineering. In this paper, we investigate the feasibility and effectiveness of translingual obfuscation with Prolog, a logic programming language. We implement translingual obfuscation in a tool called BABEL, which can selectively translate C functions into Prolog predicates. By leveraging two important features of the Prolog language, i.e., unification and backtracking, BABEL obfuscates both the data layout and control flow of C programs, making them much more difficult to reverse engineer. Our experiments show that BABEL provides effective and stealthy software obfuscation, while the cost is only modest compared to one of the most popular commercial obfuscators on the market. With BABEL, we verified the feasibility of translingual obfuscation, which we consider to be a promising new direction for software obfuscation.