Search

The use of large language models (LLMs) as automated evaluation tools to assess the quality of generated natural language, known as LLMs-as-Judges, has demonstrated promising capabilities and is rapidly gaining widespread attention. However, when applied to pairwise comparisons of candidate responses, LLM-based evaluators often exhibit selection bias. Specifically, their judgments may become inconsistent when the option positions or ID tokens are swapped, compromising the effectiveness and fairness of the evaluation result. To address this challenge, we introduce CalibraEval, a novel label-free method for mitigating selection bias during inference. Specifically, CalibraEval reformulates debiasing as an optimization task aimed at adjusting observed prediction distributions to align with unbiased prediction distributions. To solve this optimization problem, we propose a non-parametric order-preserving algorithm (NOA). This algorithm leverages the partial order relationships between model prediction distributions, thereby eliminating the need for explicit labels and precise mathematical function modeling.Empirical evaluations of LLMs in multiple representative benchmarks demonstrate that CalibraEval effectively mitigates selection bias and improves performance compared to existing debiasing methods. This work marks a step toward building more robust and unbiased automated evaluation frameworks, paving the way for improved reliability in AI-driven assessments, Comment: 13 pages

With the rapid development of large language models (LLMs), how to efficiently evaluate them has become an important research question. Existing evaluation methods often suffer from high costs, limited test formats, the need of human references, and systematic evaluation biases. To address these limitations, our study introduces the Auto-PRE, an automatic LLM evaluation framework based on peer review. In contrast to previous studies that rely on human annotations, Auto-PRE selects evaluator LLMs automatically based on their inherent traits including consistency, self-confidence, and pertinence. We conduct extensive experiments on three tasks: summary generation, non-factoid question-answering, and dialogue generation. Experimental results indicate our Auto-PRE achieves state-of-the-art performance at a lower cost. Moreover, our study highlights the impact of prompt strategies and evaluation formats on evaluation performance, offering guidance for method optimization in the future.

Benchmarking the performance of quantum error correction codes in physical systems is crucial for achieving fault-tolerant quantum computing. Current methodologies, such as (shadow) tomography or direct fidelity estimation, fall short in efficiency due to the neglect of possible prior knowledge about quantum states. To address the challenge, we introduce a framework of quantum subspace verification, employing the knowledge of quantum error correction code subspaces to reduce the potential measurement budgets. Specifically, we give the sample complexity to estimate the fidelity to the target subspace under some confidence level. Building on the framework, verification operators are developed, which can be implemented with experiment-friendly local measurements for stabilizer codes and quantum low-density parity-check (QLDPC) codes. Our constructions require $O(n-k)$ local measurement settings for both, and the sample complexity of $O(n-k)$ for stabilizer codes and of $O((n-k)^2)$ for generic QLDPC codes, where $n$ and $k$ are the numbers of physical and logical qubits, respectively. Notably, for certain codes like the notable Calderbank-Shor-Steane codes and QLDPC stabilizer codes, the setting number and sample complexity can be significantly reduced and are even independent of $n$. In addition, by combining the proposed subspace verification and direct fidelity estimation, we construct a protocol to verify the fidelity of general magic logical states with exponentially smaller sample complexity than previous methods. Our finding facilitates efficient and feasible verification of quantum error correction codes and also magical states, advancing the realization in practical quantum platforms., Comment: 15 pages, 1 figure

Machine unlearning (MU) is becoming a promising paradigm to achieve the "right to be forgotten", where the training trace of any chosen data points could be eliminated, while maintaining the model utility on general testing samples after unlearning. With the advancement of forgetting research, many fundamental open questions remain unanswered: do different samples exhibit varying levels of difficulty in being forgotten? Further, does the sequence in which samples are forgotten, determined by their respective difficulty levels, influence the performance of forgetting algorithms? In this paper, we identify key factor affecting unlearning difficulty and the performance of unlearning algorithms. We find that samples with higher privacy risks are more likely to be unlearning, indicating that the unlearning difficulty varies among different samples which motives a more precise unlearning mode. Built upon this insight, we propose a general unlearning framework, dubbed RSU, which consists of Ranking module and SeqUnlearn module., Comment: The senior authors of the draft are not fully convinced that the novelty is significant enough for this submission compared to the latest research progress in this area. Additionally, the senior authors have identified writing issues. Based on these two reasons, we have decided to withdraw the draft from arXiv

Unsupervised constituency parsers organize phrases within a sentence into a tree-shaped syntactic constituent structure that reflects the organization of sentence semantics. However, the traditional objective of maximizing sentence log-likelihood (LL) does not explicitly account for the close relationship between the constituent structure and the semantics, resulting in a weak correlation between LL values and parsing accuracy. In this paper, we introduce a novel objective for training unsupervised parsers: maximizing the information between constituent structures and sentence semantics (SemInfo). We introduce a bag-of-substrings model to represent the semantics and apply the probability-weighted information metric to estimate the SemInfo. Additionally, we develop a Tree Conditional Random Field (TreeCRF)-based model to apply the SemInfo maximization objective to Probabilistic Context-Free Grammar (PCFG) induction, the state-of-the-art method for unsupervised constituency parsing. Experiments demonstrate that SemInfo correlates more strongly with parsing accuracy than LL. Our algorithm significantly enhances parsing accuracy by an average of 7.85 points across five PCFG variants and in four languages, achieving new state-of-the-art results in three of the four languages.

Code naturalness, which captures repetitiveness and predictability in programming languages, has proven valuable for various code-related tasks in software engineering. However, precisely measuring code naturalness remains a fundamental challenge. Existing methods measure code naturalness over individual lines of code while ignoring the deep semantic relations among different lines, e.g., program dependency, which may negatively affect the precision of the measure. In this study, we aim to perform the first empirical study to investigate whether incorporating code dependency, instead of analyzing individual lines, can enhance the precision of measuring code naturalness. To achieve that, we first propose a new method named DAN for measuring code naturalness by incorporating the rich dependency information in the code. Specifically, DAN extracts multiple sequences of code lines by traversing the program dependency graph, where different code lines are connected by dependencies in each sequence, and then the code naturalness will be measured by taking each sequence as a whole. In this way, the dependency information can be well captured. Finally, we have conducted an extensive study to evaluate the influence of code dependency for measuring code naturalness with DAN, and compared it with the state-of-the-art methods under three emerging application scenarios of code naturalness. The results demonstrate that DAN can not only better distinguish natural and unnatural code, but also substantially boost two important downstream applications of code naturalness, i.e., distinguishing buggy and non-buggy code lines and data cleansing for training better code models, reflecting the significance of code dependency in measuring code naturalness.

Literature in traditional program fuzzing has confirmed that effectiveness is largely impacted by redundancy among initial seeds, thereby proposing a series of seed selection methods. JVM fuzzing, compared to traditional ones, presents unique characteristics, including large-scale and intricate code, and programs with both syntactic and semantic features. However, it remains unclear whether the existing seed selection methods are suitable for JVM fuzzing and whether utilizing program features can enhance effectiveness. To address this, we devise a total of 10 initial seed selection methods, comprising coverage-based, prefuzz-based, and program-feature-based methods. We then conduct an empirical study on three JVM implementations to extensively evaluate the performance of the seed selection methods within two SOTA fuzzing techniques (JavaTailor and VECT). Specifically, we examine performance from three aspects: (i) effectiveness and efficiency using widely studied initial seeds, (ii) effectiveness using the programs in the wild, and (iii) the ability to detect new bugs. Evaluation results first show that the program-feature-based method that utilizes the control flow graph not only has a significantly lower time overhead (i.e., 30s), but also outperforms other methods, achieving 142% to 269% improvement compared to the full set of initial seeds. Second, results reveal that the initial seed selection greatly improves the quality of wild programs and exhibits complementary effectiveness by detecting new behaviors. Third, results demonstrate that given the same testing period, initial seed selection improves the JVM fuzzing techniques by detecting more unknown bugs. Particularly, 21 out of the 25 detected bugs have been confirmed or fixed by developers. This work takes the first look at initial seed selection in JVM fuzzing, confirming its importance in fuzzing effectiveness and efficiency.

Mutation testing is vital for ensuring software quality. However, the presence of equivalent mutants is known to introduce redundant cost and bias issues, hindering the effectiveness of mutation testing in practical use. Although numerous equivalent mutant detection (EMD) techniques have been proposed, they exhibit limitations due to the scarcity of training data and challenges in generalizing to unseen mutants. Recently, large language models (LLMs) have been extensively adopted in various code-related tasks and have shown superior performance by more accurately capturing program semantics. Yet the performance of LLMs in equivalent mutant detection remains largely unclear. In this paper, we conduct an empirical study on 3,302 method-level Java mutant pairs to comprehensively investigate the effectiveness and efficiency of LLMs for equivalent mutant detection. Specifically, we assess the performance of LLMs compared to existing EMD techniques, examine the various strategies of LLMs, evaluate the orthogonality between EMD techniques, and measure the time overhead of training and inference. Our findings demonstrate that LLM-based techniques significantly outperform existing techniques (i.e., the average improvement of 35.69% in terms of F1-score), with the fine-tuned code embedding strategy being the most effective. Moreover, LLM-based techniques offer an excellent balance between cost (relatively low training and inference time) and effectiveness. Based on our findings, we further discuss the impact of model size and embedding quality, and provide several promising directions for future research. This work is the first to examine LLMs in equivalent mutant detection, affirming their effectiveness and efficiency., Comment: This paper has been accepted by ISSTA'2024

Simultaneous imaging of fluorescence-labeled and label-free phase objects in the same sample provides distinct and complementary information. Most multimodal fluorescence-phase imaging operates in transmission mode, capturing fluorescence images and phase images separately or sequentially, which limits their practical application in vivo. Here, we develop fluorescence diffraction tomography (FDT) with explicit neural fields to reconstruct the 3D refractive index (RI) of phase objects from diffracted fluorescence images captured in reflection mode. The successful reconstruction of 3D RI using FDT relies on four key components: a coarse-to-fine structure, self-calibration, a differential multi-slice rendering model, and partially coherent masks. The explicit representation integrates with the coarse-to-fine structure for high-speed, high-resolution reconstruction, while the differential multi-slice rendering model enables self-calibration of fluorescence illumination, ensuring accurate forward image prediction and RI reconstruction. Partially coherent masks efficiently resolve discrepancies between the coherent light model and partially coherent light data. FDT successfully reconstructs the RI of 3D cultured label-free bovine myotubes in a 530 $\times$ 530 $\times$ 300 $\mu m^3$ volume at 1024 $\times$ 1024 pixels across 24 $z$-layers from fluorescence images, demonstrating high resolution and high accuracy 3D RI reconstruction of bulky and heterogeneous biological samples in vitro.

Deep Learning (DL) compilers typically load a DL model and optimize it with intermediate representation.Existing DL compiler testing techniques mainly focus on model optimization stages, but rarely explore bug detection at the model loading stage. Effectively testing the model loading stage requires covering diverse usages of each DL operator from various DL libraries, which shares a common objective with DL library testing, indicating that the embedded knowledge in DL library tests is beneficial for testing the model loading stage of DL compilers. In this work, we propose OPERA to extract such domain knowledge from the test inputs for DL libraries. OPERA constructs diverse tests from the various test inputs for DL libraries (including the test inputs documented in DL libraries and those generated by recent fuzzers). In addition, it incorporates a diversity-based test prioritization strategy to migrate and execute those test inputs that are more likely to detect diverse bugs earlier. We considered three sources of tests in DL libraries for migration and used eight frontends from three DL compilers (e.g., TVM, TensorRT, and OpenVINO) for evaluation. OPERA detected 170 previously unknown bugs in total, 90 of which have been confirmed/fixed by developers, demonstrating the effectiveness of such the migration-based idea. The test prioritization strategy in OPERA improves testing efficiency with migrated tests by 11.9%~47.4% on average compared to general test prioritization strategies., Comment: This paper has been accepted by ICSE'2025

The continual evolution of autonomous driving technology requires car-following models that can adapt to diverse and dynamic traffic environments. Traditional learning-based models often suffer from performance degradation when encountering unseen traffic patterns due to a lack of continual learning capabilities. This paper proposes a novel car-following model based on continual learning that addresses this limitation. Our framework incorporates Elastic Weight Consolidation (EWC) and Memory Aware Synapses (MAS) techniques to mitigate catastrophic forgetting and enable the model to learn incrementally from new traffic data streams. We evaluate the performance of the proposed model on the Waymo and Lyft datasets which encompass various traffic scenarios. The results demonstrate that the continual learning techniques significantly outperform the baseline model, achieving 0\% collision rates across all traffic conditions. This research contributes to the advancement of autonomous driving technology by fostering the development of more robust and adaptable car-following models.

Accurate modeling of car-following behaviors is essential for various applications in traffic management and autonomous driving systems. However, current approaches often suffer from limitations like high sensitivity to data quality and lack of interpretability. In this study, we propose GenFollower, a novel zero-shot prompting approach that leverages large language models (LLMs) to address these challenges. We reframe car-following behavior as a language modeling problem and integrate heterogeneous inputs into structured prompts for LLMs. This approach achieves improved prediction performance and interpretability compared to traditional baseline models. Experiments on the Waymo Open datasets demonstrate GenFollower's superior performance and ability to provide interpretable insights into factors influencing car-following behavior. This work contributes to advancing the understanding and prediction of car-following behaviors, paving the way for enhanced traffic management and autonomous driving systems.

Solidity compiler plays a key role in enabling the development of smart contract applications on Ethereum by governing the syntax of a domain-specific language called Solidity and performing compilation and optimization of Solidity code. The correctness of Solidity compiler is critical in fostering transparency, efficiency, and trust in industries reliant on smart contracts. However, like other software systems, Solidity compiler is prone to bugs, which may produce incorrect bytecodes on blockchain platforms, resulting in severe security concerns. As a domain-specific compiler for smart contracts, Solidity compiler differs from other compilers in many perspectives, posing unique challenges to detect its bugs. To understand the bugs in Solidity compiler and benefit future research, in this paper, we present the first systematic study on 533 Solidity compiler bugs. We carefully examined their characteristics (including symptoms, root causes, and distribution), and their triggering test cases. Our study leads to seven bug-revealing takeaways for Solidity compiler. Moreover, to study the limitations of Solidity compiler fuzzers and bring our findings into practical scenarios, we evaluate three Solidity compiler fuzzers on our constructed benchmark. The results show that these fuzzers are inefficient in detecting Solidity compiler bugs. The inefficiency arises from their failure to consider the interesting bug-inducing features, bug-related compilation flags, and test oracles

In this paper, we initiate the computational problem of jointly designing information and contracts. We consider three possible classes of contracts with decreasing flexibility and increasing simplicity: ambiguous contracts, menus of explicit contracts and explicit single contract. Ambiguous contracts allow the principal to conceal the applied payment schemes through a contract that depends on the unknown state of nature, while explicit contracts reveal the contract prior to the agent's decision. Our results show a trade-off between the simplicity of the contracts and the computational complexity of the joint design. Indeed, we show that an approximately-optimal mechanism with ambiguous contracts can be computed in polynomial time. However, they are convoluted mechanisms and not well-suited for some real-world scenarios. Conversely, explicit menus of contracts and single contracts are simpler mechanisms, but they cannot be computed efficiently. In particular, we show that computing the optimal mechanism with explicit menus of contracts and single contracts is APX-Hard. We also characterize the structure of optimal mechanisms. Interestingly, direct mechanisms are optimal for both the most flexible ambiguous contracts and the least flexible explicit single contract, but they are suboptimal for that with menus of contracts. Finally, motivated by our hardness results, we turn our attention to menus of linear contracts and single linear contracts. We show that both the problem of computing the optimal mechanism with an explicit menu of linear contracts and an explicit single linear contract admits an FPTAS., Comment: contract design, information design

Unit testing is an essential activity in software development for verifying the correctness of software components. However, manually writing unit tests is challenging and time-consuming. The emergence of Large Language Models (LLMs) offers a new direction for automating unit test generation. Existing research primarily focuses on closed-source LLMs (e.g., ChatGPT and CodeX) with fixed prompting strategies, leaving the capabilities of advanced open-source LLMs with various prompting settings unexplored. Particularly, open-source LLMs offer advantages in data privacy protection and have demonstrated superior performance in some tasks. Moreover, effective prompting is crucial for maximizing LLMs' capabilities. In this paper, we conduct the first empirical study to fill this gap, based on 17 Java projects, five widely-used open-source LLMs with different structures and parameter sizes, and comprehensive evaluation metrics. Our findings highlight the significant influence of various prompt factors, show the performance of open-source LLMs compared to the commercial GPT-4 and the traditional Evosuite, and identify limitations in LLM-based unit test generation. We then derive a series of implications from our study to guide future research and practical use of LLM-based unit test generation., Comment: Accepted by ASE 2024, Research Paper Track

The prevalence of sarcasm in social media, conveyed through text-image combinations, presents significant challenges for sentiment analysis and intention mining. Existing multi-modal sarcasm detection methods have been proven to overestimate performance, as they struggle to effectively capture the intricate sarcastic cues that arise from the interaction between an image and text. To address these issues, we propose InterCLIP-MEP, a novel framework for multi-modal sarcasm detection. Specifically, we introduce an Interactive CLIP (InterCLIP) as the backbone to extract text-image representations, enhancing them by embedding cross-modality information directly within each encoder, thereby improving the representations to capture text-image interactions better. Furthermore, an efficient training strategy is designed to adapt InterCLIP for our proposed Memory-Enhanced Predictor (MEP). MEP uses a dynamic, fixed-length dual-channel memory to store historical knowledge of valuable test samples during inference. It then leverages this memory as a non-parametric classifier to derive the final prediction, offering a more robust recognition of multi-modal sarcasm. Experiments demonstrate that InterCLIP-MEP achieves state-of-the-art performance on the MMSD2.0 benchmark, with an accuracy improvement of 1.08% and an F1 score improvement of 1.51% over the previous best method., Comment: 9 pages, 6 figures, 3 tables; Code and data are available at https://github.com/CoderChen01/InterCLIP-MEP

Photo stimulated luminescent materials are one of the most attractive alternatives for next generation optical information storage technologies. However, there are still some challenges in regulating appropriate energy levels in luminescent materials for optical information storage. Herein, a green emission nanophosphor Zn2SiO4: Cr3+, Mn2+ with the trap depth of 1.05 eV, fulfilling the requirements for optical information storage, was fabricated for the first time through the solution combustion method and subsequent heat treatment at 1000 degree centigrade for 2h. The crystal structure, micromorphology, photoluminescence (PL), photoluminescence excitation (PLE), and afterglow properties of Zn2SiO4: xCr3+, yMn2+ were studied systematically. By applying the strategy of trap depth engineering, high trap density with proper trap depth was observed when Cr3+ ions were introduced into Zn2SiO4: Mn2+. Thermoluminescence (TL) glow curve analysis through the initial rise (IR) method was conducted to gain some insight into the information of traps. As proof of application, information storage was experimentally achieved by choosing 275 nm illumination for information writing and 980 nm NIR excitation for information reading. The results indicate that Zn2SiO4: Cr3+, Mn2+ phosphor holds promise for potential applications in the field of optical information storage.

Unsupervised constituency parsing focuses on identifying word sequences that form a syntactic unit (i.e., constituents) in target sentences. Linguists identify the constituent by evaluating a set of Predicate-Argument Structure (PAS) equivalent sentences where we find the constituent appears more frequently than non-constituents (i.e., the constituent corresponds to a frequent word sequence within the sentence set). However, such frequency information is unavailable in previous parsing methods that identify the constituent by observing sentences with diverse PAS. In this study, we empirically show that constituents correspond to frequent word sequences in the PAS-equivalent sentence set. We propose a frequency-based parser span-overlap that (1) computes the span-overlap score as the word sequence's frequency in the PAS-equivalent sentence set and (2) identifies the constituent structure by finding a constituent tree with the maximum span-overlap score. The parser achieves state-of-the-art level parsing accuracy, outperforming existing unsupervised parsers in eight out of ten languages. Additionally, we discover a multilingual phenomenon: participant-denoting constituents tend to have higher span-overlap scores than equal-length event-denoting constituents, meaning that the former tend to appear more frequently in the PAS-equivalent sentence set than the latter. The phenomenon indicates a statistical difference between the two constituent types, laying the foundation for future labeled unsupervised parsing research.

Automatic test generation plays a critical role in software quality assurance. While the recent advances in Search-Based Software Testing (SBST) and Large Language Models (LLMs) have shown promise in generating useful tests, these techniques still struggle to cover certain branches. Reaching these hard-to-cover branches usually requires constructing complex objects and resolving intricate inter-procedural dependencies in branch conditions, which poses significant challenges for existing test generation techniques. In this work, we propose TELPA, a novel technique aimed at addressing these challenges. Its key insight lies in extracting real usage scenarios of the target method under test to learn how to construct complex objects and extracting methods entailing inter-procedural dependencies with hard-to-cover branches to learn the semantics of branch constraints. To enhance efficiency and effectiveness, TELPA identifies a set of ineffective tests as counter-examples for LLMs and employs a feedback-based process to iteratively refine these counter-examples. Then, TELPA integrates program analysis results and counter-examples into the prompt, guiding LLMs to gain deeper understandings of the semantics of the target method and generate diverse tests that can reach the hard-to-cover branches. Our experimental results on 27 open-source Python projects demonstrate that TELPA significantly outperforms the state-of-the-art SBST and LLM-based techniques, achieving an average improvement of 31.39% and 22.22% in terms of branch coverage., Comment: 11 pages, 4 figures

The main result of this paper is an almost approximation-preserving polynomial-time reduction from the most general multi-parameter Bayesian contract design (BCD) to single-parameter BCD. That is, for any multi-parameter BCD instance $I^M$, we construct a single-parameter instance $I^S$ such that any $\beta$-approximate contract (resp. menu of contracts) of $I^S$ can in turn be converted to a $(\beta -\epsilon)$-approximate contract (resp. menu of contracts) of $I^M$. The reduction is in time polynomial in the input size and $\log(\frac{1}{\epsilon})$; moreover, when $\beta = 1$ (i.e., the given single-parameter solution is exactly optimal), the dependence on $\frac{1}{\epsilon}$ can be removed, leading to a polynomial-time exact reduction. This efficient reduction is somewhat surprising because in the closely related problem of Bayesian mechanism design, a polynomial-time reduction from multi-parameter to single-parameter setting is believed to not exist. Our result demonstrates the intrinsic difficulty of addressing moral hazard in Bayesian contract design, regardless of being single-parameter or multi-parameter. As byproducts, our reduction answers two open questions in recent literature of algorithmic contract design: (a) it implies that optimal contract design in single-parameter BCD is not in APX unless P=NP even when the agent's type distribution is regular, answering the open question of [Alon et al. 2021] in the negative; (b) it implies that the principal's (order-wise) tight utility gap between using a menu of contracts and a single contract is $\Theta(n)$ where $n$ is the number of actions, answering the major open question of [Guruganesh et al. 2021] for the single-parameter case., Comment: update some results

Category-agnostic pose estimation (CAPE) aims to predict keypoints for arbitrary classes given a few support images annotated with keypoints. Existing methods only rely on the features extracted at support keypoints to predict or refine the keypoints on query image, but a few support feature vectors are local and inadequate for CAPE. Considering that human can quickly perceive potential keypoints of arbitrary objects, we propose a novel framework for CAPE based on such potential keypoints (named as meta-points). Specifically, we maintain learnable embeddings to capture inherent information of various keypoints, which interact with image feature maps to produce meta-points without any support. The produced meta-points could serve as meaningful potential keypoints for CAPE. Due to the inevitable gap between inherency and annotation, we finally utilize the identities and details offered by support keypoints to assign and refine meta-points to desired keypoints in query image. In addition, we propose a progressive deformable point decoder and a slacked regression loss for better prediction and supervision. Our novel framework not only reveals the inherency of keypoints but also outperforms existing methods of CAPE. Comprehensive experiments and in-depth studies on large-scale MP-100 dataset demonstrate the effectiveness of our framework., Comment: Published in CVPR 2024

In this paper, we introduce I3DE (Inconsistency Inspecting IDE) - an IDE plugin to inspect inconsistencies in PL/SQL code. We first observed the potential issues, e.g., misuses or bugs, that are introduced by the inconsistent understanding of PL/SQL semantics by PL/SQL programmers and DBMS developers, and propose a metamorphic testing-based approach for inspecting such inconsistencies in PL/SQL code. We design and implement our approach in I3DE, a widely usable plugin for the IntelliJ Platform. We conducted a comparative user study involving 16 participants, and the findings indicate that I3DE is consistently effective and efficient in helping programmers identify and avoid inconsistencies across different programming difficulties, Comment: In 2024 First IDE Workshop (IDE' 24), April 20, 2024, Lisbon, Portugal. ACM, New York, NY, USA, 6 pages

Fairness has been a critical issue that affects the adoption of deep learning models in real practice. To improve model fairness, many existing methods have been proposed and evaluated to be effective in their own contexts. However, there is still no systematic evaluation among them for a comprehensive comparison under the same context, which makes it hard to understand the performance distinction among them, hindering the research progress and practical adoption of them. To fill this gap, this paper endeavours to conduct the first large-scale empirical study to comprehensively compare the performance of existing state-of-the-art fairness improving techniques. Specifically, we target the widely-used application scenario of image classification, and utilized three different datasets and five commonly-used performance metrics to assess in total 13 methods from diverse categories. Our findings reveal substantial variations in the performance of each method across different datasets and sensitive attributes, indicating over-fitting on specific datasets by many existing methods. Furthermore, different fairness evaluation metrics, due to their distinct focuses, yield significantly different assessment results. Overall, we observe that pre-processing methods and in-processing methods outperform post-processing methods, with pre-processing methods exhibiting the best performance. Our empirical study offers comprehensive recommendations for enhancing fairness in deep learning models. We approach the problem from multiple dimensions, aiming to provide a uniform evaluation platform and inspire researchers to explore more effective fairness solutions via a set of implications., Comment: Accepted by the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2024). Please include ISSTA in any citations

Redundancy-based automated program repair (APR), which generates patches by referencing existing source code, has gained much attention since they are effective in repairing real-world bugs with good interpretability. However, since existing approaches either demand the existence of multi-line similar code or randomly reference existing code, they can only repair a small number of bugs with many incorrect patches, hindering their wide application in practice. In this work, we aim to improve the effectiveness of redundancy-based APR by exploring more effective source code reuse methods for improving the number of correct patches and reducing incorrect patches. Specifically, we have proposed a new repair technique named Repatt, which incorporates a two-level pattern mining process for guiding effective patch generation (i.e., token and expression levels). We have conducted an extensive experiment on the widely-used Defects4J benchmark and compared Repatt with eight state-of-the-art APR approaches. The results show that our approach complements existing approaches by repairing {15} unique bugs compared with the latest deep learning-based methods and {19} unique bugs compared with traditional repair methods when providing the perfect fault localization. In addition, when the perfect fault localization is unknown in real practice, Repatt significantly outperforms the baseline approaches by achieving much higher patch precision, i.e., {83.8\%}. Moreover, we further proposed an effective patch ranking strategy for combining the strength of Repatt and the baseline methods. The result shows that it repairs 124 bugs when only considering the Top-1 patches and improves the best-performing repair method by repairing 39 more bugs. The results demonstrate the effectiveness of our approach for practical use.

Code generation is to automatically generate source code conforming to a given programming specification, which has received extensive attention especially with the development of large language models (LLMs). Due to the inherent difficulty of code generation, the code generated by LLMs may be not aligned with the specification. To improve the performance of LLMs in code generation, some thought-eliciting prompting techniques have been proposed to guide LLMs for specification understanding. However, it is still hard to produce correct understanding for complicated programming problems, leading to unsatisfactory code generation performance. Also, some feedback-based prompting techniques have been proposed to fix incorrect code using error messages produced by test execution. However, when the generated code deviates significantly from the ground truth, they encounter difficulties in improving performance based on such coarse-grained information. In this work, we propose a novel prompting technique, called {\mu}FiX, to improve the code generation performance of LLMs by devising both sophisticated thought-eliciting prompting and feedback-based prompting and making the first exploration on their synergy. It first exploits test case analysis to obtain specification understanding and enables a self-improvement process to identify and fix the misunderstanding in the thought-eliciting prompting phase. {\mu}FiX further fixes the specification understanding towards the direction reducing the gap between the provided understanding and the actual understanding implicitly utilized by LLMs for code generation in the feedback-based prompting phase. By obtaining as correct understanding as possible with {\mu}FiX, the code generation performance of LLMs can be largely improved.

To alleviate the expensive human labeling, semi-supervised semantic segmentation employs a few labeled images and an abundant of unlabeled images to predict the pixel-level label map with the same size. Previous methods often adopt co-training using two convolutional networks with the same architecture but different initialization, which fails to capture the sufficiently diverse features. This motivates us to use tri-training and develop the triple-view encoder to utilize the encoders with different architectures to derive diverse features, and exploit the knowledge distillation skill to learn the complementary semantics among these encoders. Moreover, existing methods simply concatenate the features from both encoder and decoder, resulting in redundant features that require large memory cost. This inspires us to devise a dual-frequency decoder that selects those important features by projecting the features from the spatial domain to the frequency domain, where the dual-frequency channel attention mechanism is introduced to model the feature importance. Therefore, we propose a Triple-view Knowledge Distillation framework, termed TriKD, for semi-supervised semantic segmentation, including the triple-view encoder and the dual-frequency decoder. Extensive experiments were conducted on two benchmarks, \ie, Pascal VOC 2012 and Cityscapes, whose results verify the superiority of the proposed method with a good tradeoff between precision and inference speed.

Multi-task learning (MTL) aims to improve the performance of a primary task by jointly learning with related auxiliary tasks. Traditional MTL methods select tasks randomly during training. However, both previous studies and our results suggest that such a random selection of tasks may not be helpful, and can even be harmful to performance. Therefore, new strategies for task selection and assignment in MTL need to be explored. This paper studies the multi-modal, multi-task dialogue act classification task, and proposes a method for selecting and assigning tasks based on non-stationary multi-armed bandits (MAB) with discounted Thompson Sampling (TS) using Gaussian priors. Our experimental results show that in different training stages, different tasks have different utility. Our proposed method can effectively identify the task utility, actively avoid useless or harmful tasks, and realise the task assignment during training. Our proposed method is significantly superior in terms of UAR and F1 to the single-task and multi-task baselines with p-values < 0.05. Further analysis of experiments indicates that for the dataset with the data imbalance problem, our proposed method has significantly higher stability and can obtain consistent and decent performance for minority classes. Our proposed method is superior to the current state-of-the-art model., Comment: Accepted by ICASSP 2024

The development of correct and efficient software can be hindered by compilation errors, which must be fixed to ensure the code's syntactic correctness and program language constraints. Neural network-based approaches have been used to tackle this problem, but they lack guarantees of output correctness and can require an unlimited number of modifications. Fixing compilation errors within a given number of modifications is a challenging task. We demonstrate that finding the minimum number of modifications to fix a compilation error is NP-hard. To address compilation error fixing problem, we propose OrdinalFix, a complete algorithm based on shortest-path CFL (context-free language) reachability with attribute checking that is guaranteed to output a program with the minimum number of modifications required. Specifically, OrdinalFix searches possible fixes from the smallest to the largest number of modifications. By incorporating merged attribute checking to enhance efficiency, the time complexity of OrdinalFix is acceptable for application. We evaluate OrdinalFix on two datasets and demonstrate its ability to fix compilation errors within reasonable time limit. Comparing with existing approaches, OrdinalFix achieves a success rate of 83.5%, surpassing all existing approaches (71.7%)., Comment: Accepted by ASE 2023

Semantic communications (SC) have been expected to be a new paradigm shifting to catalyze the next generation communication, whose main concerns shift from accurate bit transmission to effective semantic information exchange in communications. However, the previous and widely-used metrics for images are not applicable to evaluate the image semantic similarity in SC. Classical metrics to measure the similarity between two images usually rely on the pixel level or the structural level, such as the PSNR and the MS-SSIM. Straightforwardly using some tailored metrics based on deep-learning methods in CV community, such as the LPIPS, is infeasible for SC. To tackle this, inspired by BERTScore in NLP community, we propose a novel metric for evaluating image semantic similarity, named Vision Transformer Score (ViTScore). We prove theoretically that ViTScore has 3 important properties, including symmetry, boundedness, and normalization, which make ViTScore convenient and intuitive for image measurement. To evaluate the performance of ViTScore, we compare ViTScore with 3 typical metrics (PSNR, MS-SSIM, and LPIPS) through 4 classes of experiments: (i) correlation with BERTScore through evaluation of image caption downstream CV task, (ii) evaluation in classical image communications, (iii) evaluation in image semantic communication systems, and (iv) evaluation in image semantic communication systems with semantic attack. Experimental results demonstrate that ViTScore is robust and efficient in evaluating the semantic similarity of images. Particularly, ViTScore outperforms the other 3 typical metrics in evaluating the image semantic changes by semantic attack, such as image inverse with Generative Adversarial Networks (GANs). This indicates that ViTScore is an effective performance metric when deployed in SC scenarios.