We introduce DSCodeBench, a new benchmark designed to evaluate large language models (LLMs) on complicated and realistic data science code generation tasks. DSCodeBench consists of 1,000 carefully constructed problems sourced from realistic problems from GitHub across ten widely used Python data science libraries. DSCodeBench offers a more challenging and representative testbed, more complex code solutions, more comprehensive data science libraries, clearer and better structured problem descriptions, and stronger test suites. To construct the DSCodeBench, we develop a robust pipeline that combines task scope selection, code construction, test case generation, and problem description synthesis. The process is paired with rigorous manual editing to ensure alignment and enhance the reliability of the evaluation. Experimental result shows that DSCodeBench exhibits robust scaling behavior, where larger models systematically outperform smaller ones, validating its ability to distinguish model capabilities. The best LLM we test, GPT-4o, has a pass@1 of 0.392, indicating that LLMs still have a large room to improve for realistic data science code generation tasks. We believe DSCodeBench will serve as a rigorous and trustworthy foundation for advancing LLM-based data science programming.

While Large Language Models (LLMs) have substantially improved the functional correctness of code translation, the critical dimension of execution efficiency remains overlooked.We present TRACE, the first benchmark to explicitly assess efficiency in LLM-translated code.TRACE includes 1,000 efficiency-critical tasks across C++, Java, and Python, each augmented with stress tests that reveal efficiency disparities often overlooked by small-scale tests.Using TRACE, we conduct an extensive evaluation of 28 representative LLMs and highlight several key insights: 1) Correctness and efficiency are often misaligned: the correctness leader Claude-Sonnet-4-Think achieves only moderate time efficiency, outperformed by smaller open-source LLMs such as Qwen2.5-Coder-14B-Instruct.2) Inefficiency is both prevalent and patterned: 23.5% of correct translations suffer from notable inefficiency, mainly arising from algorithm implementation discrepancy (11.9%), language construct mismatch (66.4%), and resource management inefficiency (21.7%).3) Inference-time prompt strategies bring only modest improvements, indicating that simple prompting alone is insufficient to improve translation efficiency.Together, our results establish execution efficiency as an essential dimension of code translation and position TRACE as a principled foundation for efficiency-oriented evaluation.Our

Time series forecasting requires architectures that simultaneously achieve three competing objectives: (1) strict temporal causality for reliable predictions, (2) sub-quadratic complexity for practical scalability, and (3) multi-scale pattern recognition for accurate long-horizon forecasting. We introduce AutoHFormer, a hierarchical autoregressive transformer that addresses these challenges through three key innovations: 1) Hierarchical Temporal Modeling: Our architecture decomposes predictions into segment-level blocks processed in parallel, followed by intra-segment sequential refinement. This dual-scale approach maintains temporal coherence while enabling efficient computation. 2) Dynamic Windowed Attention: The attention mechanism employs learnable causal windows with exponential decay, reducing complexity while preserving precise temporal relationships. This design avoids both the anti-causal violations of standard transformers and the sequential bottlenecks of RNN hybrids. 3) Adaptive Temporal Encoding: a novel position encoding system is adopted to capture time patterns at multiple scales. It combines fixed oscillating patterns for short-term variations with learnable decay rates for long-term trends. Comprehensive experiments demonstrate that AutoHFormer 10.76X faster training and 6.06X memory reduction compared to PatchTST on PEMS08, while maintaining consistent accuracy across 96-720 step horizons in most of cases. These breakthroughs establish new benchmarks for efficient and precise time series modeling. Implementations of our method and all baselines in hierarchical autoregressive mechanism are available at https://github.com/lizzyhku/Autotime.

Recently, deep reinforcement learning (DRL) has emerged as a promising approach for robotic control. However, the deployment of DRL in real-world robots is hindered by its sensitivity to environmental perturbations. While existing whitebox adversarial attacks rely on local gradient information and apply uniform perturbations across all states to evaluate DRL robustness, they fail to account for temporal dynamics and statespecific vulnerabilities. To combat the above challenge, we first conduct a theoretical analysis of white-box attacks in DRL by establishing the adversarial victim-dynamics Markov decision process (AVD-MDP), to derive the necessary and sufficient conditions for a successful attack. Based on this, we propose a selective state-aware reinforcement adversarial attack method, named STAR, to optimize perturbation stealthiness and state visitation dispersion. STAR first employs a soft mask-based state-targeting mechanism to minimize redundant perturbations, enhancing stealthiness and attack effectiveness. Then, it incorporates an information-theoretic optimization objective to maximize mutual information between perturbations, environmental states, and victim actions, ensuring a dispersed state-visitation distribution that steers the victim agent into vulnerable states for maximum return reduction. Extensive experiments demonstrate that STAR outperforms state-of-the-art benchmarks

As the adoption of LLMs becomes more widespread in software coding ecosystems, a pressing issue has emerged: does the generated code contain social bias and unfairness, such as those related to age, gender, and race? This issue concerns the integrity, fairness, and ethical foundation of software applications that depend on the code generated by these models but are underexplored in the literature. This article presents a novel bias testing framework that is specifically designed for code generation tasks. Based on this framework, we conduct an extensive empirical study on the biases in code generated by five widely studied LLMs (i.e., PALM-2-CodeChat-bison, Claude-instant-1, GPT-3.5-turbo, GPT-4-turbo, and GPT-4). Our findings reveal that biases are prevalent. For example, 13.47% to 49.10% of the codes generated by these LLMs have biased behaviors towards gender. Moreover, we study five bias mitigation prompt strategies that are commonly used in current code generation scenarios, i.e., zero-shot, one-shot, few-shot, and two Chain-of-Thought (CoT) prompts, with and without provided feedback-driven refinement. Our evaluation results illustrate that using direct prompt engineering strategies has limited effectiveness in mitigating bias, but our test execution feedback can help to reduce the ratio of code biases to a large extent (e.g., from 59.88% to 4.79% for GPT-4).

Guichao Zhu, Lintian Lei, Yuhao Qing, Yichao Fu, Fanxin Li, Dong Huang, Zekai Sun, Heming Cui. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Spatial-temporal graph representations play a crucial role in urban sensing applications, including traffic analysis, human mobility behavior modeling, and citywide crime prediction. However, a key challenge lies in the noisy and sparse nature of spatial-temporal data, which limits existing neural networks' ability to learn meaningful region representations in the spatial-temporal graph. To overcome these limitations, we propose HGAurban, a novel heterogeneous spatial-temporal graph masked autoencoder that leverages generative self-supervised learning for robust urban data representation. Our framework introduces a spatial-temporal heterogeneous graph encoder that extracts region-wise dependencies from multi-source data, enabling comprehensive modeling of diverse spatial relationships. Within our self-supervised learning paradigm, we implement a masked autoencoder that jointly processes node features and graph structure. This approach automatically learns heterogeneous spatial-temporal patterns across regions, significantly improving the representation of dynamic temporal correlations. Comprehensive experiments across multiple spatiotemporal mining tasks demonstrate that our framework outperforms state-of-the-art methods and robustly handles real-world urban data challenges, including noise and sparsity in both spatial and temporal dimensions.

Deep Reinforcement Learning (DRL) suffers from uncertainties and inaccuracies in the observation signal in real-world applications. Adversarial attack is an effective method for evaluating the robustness of DRL agents. However, existing attack methods targeting individual sampled actions have limited impacts on the overall policy distribution, particularly in continuous action spaces. To address these limitations, we propose the Distribution-Aware Projected Gradient Descent attack (DAPGD). DAPGD uses distribution similarity as the gradient perturbation input to attack the policy network, which leverages the entire policy distribution rather than relying on individual samples. We utilize the Bhattacharyya distance in DAPGD to measure policy similarity, enabling sensitive detection of subtle but critical differences between probability distributions. Our experiment results demonstrate that DAPGD achieves SOTA results compared to the baselines in three robot navigation tasks, achieving an average 22.03% higher reward drop compared to the best baseline.

As large language models (LLMs) play an increasingly important role in code generation, enhancing both correctness and efficiency has become crucial. Current methods primarily focus on correctness, often overlooking efficiency. To address this gap, we introduce SWIFTCODE to improve both aspects by fine-tuning LLMs on a high-quality dataset comprising correct and efficient code samples. Our methodology involves leveraging multiple LLMs to generate diverse candidate code solutions for various tasks across different programming languages. We then evaluate these solutions by directly measuring their execution time and memory usage through local execution. The code solution with the lowest execution time and memory consumption is selected as the final output for each task. Experimental results demonstrate significant improvements when fine-tuning with SWIFTCODE. For instance, Qwen2.5-Coder-7B-Instruct’s pass@1 score increases from 44.8% to 57.7%, while the average execution time for correct tasks decreases by 48.4%. SWIFTCODE offers a scalable and effective solution for advancing AI-driven code generation, benefiting both software development and computational problem-solving.

Deep neural networks (DNNs) are vulnerable to adversarial examples, in which DNNs are misled to false outputs due to inputs containing imperceptible perturbations. Adversarial training, a reliable and effective method of defense, may significantly reduce the vulnerability of neural networks and becomes the de facto standard for robust learning. While many recent works practice the data-centric philosophy, such as how to generate better adversarial examples or use generative models to produce additional training data, we look back to the models themselves and revisit the adversarial robustness from the perspective of deep feature distribution as an insightful complementarity. In this paper, we propose Branch Orthogonality adveRsarial Training (BORT) to obtain state-of-the-art performance with solely the original dataset for adversarial training. To practice our design idea of integrating multiple orthogonal solution spaces, we leverage a simple multi-branch neural network and propose a corresponding loss function, branch-orthogonal loss, to make each solution space of the multi-branch model orthogonal. We evaluate our approach on CIFAR-10, CIFAR-100 and SVHN against$\ell_{\infty}$norm-bounded perturbations of size$\epsilon=8 / 255$, respectively. Exhaustive experiments are conducted to show that our method goes beyond all state-of-the-art methods without any tricks. Compared to all methods that do not use additional data for training, our models achieve 67.3 % and 41.5 % robust accuracy on CIFAR-10 and CIFAR-100 (improving upon the state-of-the-art by$\mathbf{+7.23 \%}$and$\mathbf{+9.07 \%}$).

Deep reinforcement learning (DRL) has emerged as a promising approach for robotic control, but its real-world deployment remains challenging due to its vulnerability to environmental perturbations. Existing white-box adversarial attack methods, adapted from supervised learning, fail to effectively target DRL agents as they overlook temporal dynamics and indiscriminately perturb all state dimensions, limiting their impact on long-term rewards. To address these challenges, we propose the Adaptive Gradient-Masked Reinforcement (AGMR) Attack, a white-box attack method that combines DRL with a gradient-based soft masking mechanism to dynamically identify critical state dimensions and optimize adversarial policies. AGMR selectively allocates perturbations to the most impactful state features and incorporates a dynamic adjustment mechanism to balance exploration and exploitation during training. Extensive experiments demonstrate that AGMR outperforms state-of-the-art adversarial attack methods in degrading the performance of the victim agent and enhances the victim agent’s robustness through adversarial defense mechanisms.

Large Language Models (LLMs) generate functionally correct solutions but often fall short in code efficiency, a critical bottleneck for real-world deployment. In this paper, we introduce a novel test-time iterative optimization framework to address this, employing a closed-loop system where LLMs iteratively refine code based on empirical performance feedback from an execution sandbox. We explore three training strategies: Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Group Relative Policy Optimization~(GRPO). Experiments on our Venus dataset and the APPS benchmark show that SFT and DPO rapidly saturate in efficiency gains. In contrast, GRPO, using reinforcement learning (RL) with execution feedback, continuously optimizes code performance, significantly boosting both pass@1 (from 47% to 62%) and the likelihood of outperforming human submissions in efficiency (from 31% to 45%). Our work demonstrates effective test-time code efficiency improvement and critically reveals the power of RL in teaching LLMs to truly self-improve code efficiency. We released our code and data at https://github.com/Elfsong/Afterburner.

Existing code generation benchmarks primarily evaluate functional correctness, with limited attention to code efficiency, and they are often restricted to a single language such as Python. To address this gap, we introduce EffiBench‑X, the first large‑scale multi‑language benchmark specifically designed for robust efficiency evaluation of LLM‑generated code. EffiBench‑X supports Python, C++, Java, JavaScript, Ruby, and Go, and comprises competitive programming tasks paired with human‑expert solutions as efficiency baselines. Evaluating state‑of‑the‑art LLMs on EffiBench‑X reveals that while models frequently generate functionally correct code, they consistently underperform human experts in efficiency. Even the most efficient LLM‑generated solutions (e.g., Qwen3‑32B) achieve only around 62% of human efficiency on average, with significant language‑specific variation: models tend to perform better in Python, Ruby, and JavaScript than in Java, C++, and Go (e.g., DeepSeek‑R1’s Python code is markedly more efficient than its Java code). These findings highlight the need for research into optimization‑oriented methods to improve the efficiency of LLM‑generated code across diverse languages. The dataset and evaluation infrastructure are publicly available at https://github.com/EffiBench/EffiBench-X.git and https://huggingface.co/datasets/EffiBench/effibench-x.

Deep neural networks have been widely used in many critical applications, such as autonomous vehicles and medical diagnosis. However, their security is threatened by backdoor attacks, which are achieved by adding artificial patterns to specific training data. Existing defense strategies primarily focus on using reverse engineering to reproduce the backdoor trigger generated by attackers and subsequently repair the DNN model by adding the trigger into inputs and fine-tuning the model with ground truth labels. However, once the trigger generated by the attackers is complex and invisible, the defender cannot reproduce the trigger successfully then the DNN model will not be repaired, as the trigger is not effectively removed. In this work, we propose Adversarial Feature Map Pruning for Backdoor (FMP) to mitigate backdoor from the DNN. Unlike existing defense strategies, which focus on reproducing backdoor triggers, FMP attempts to prune backdoor feature maps, which are trained to extract backdoor information from inputs. After pruning these backdoor feature maps, FMP will fine-tune the model with a secure subset of training data. Our experiments demonstrate that, compared to existing defense strategies, FMP can effectively reduce the Attack Success Rate (ASR) even against the most complex and invisible attack triggers (e.g., FMP decreases the ASR to 2.86% in CIFAR10, which is 19.2% to 65.41% lower than baselines). Second, unlike conventional defense methods that tend to exhibit low robust accuracy (that is, the accuracy of the model on poisoned data), FMP achieves a higher RA, indicating its superiority in maintaining model performance while mitigating the effects of backdoor attacks (e.g., FMP obtains 87.40% RA in CIFAR10). Our code is publicly available at: https://github.com/hku-systems/FMP.

The exceptional mobility and long endurance of air-ground robots are raising interest in their usage to navigate complex environments (e.g., forests and large buildings). However, such environments often contain occluded and unknown regions, and without accurate prediction of unobserved obstacles, the movement of the air-ground robot often suffers a sub-optimal trajectory under existing mapping-based and learning-based navigation methods. In this work, we present AGRNav, a novel framework designed to search for safe and energy-saving air-ground hybrid paths. AGRNav contains a lightweight semantic scene completion network (SCONet) with self-attention to enable accurate obstacle predictions by capturing contextual information and occlusion area features. The framework subsequently employs a query-based method for low-latency updates of prediction results to the grid map. Finally, based on the updated map, the hierarchical path planner efficiently searches for energy-saving paths for navigation. We validate AGRNav’s performance through benchmarks in both simulated and real-world environments, demonstrating its superiority over classical and state-of-the-art methods. The open-source code is available at https://github.com/jmwang0117/AGRNav.

Deep Learning Systems (DLSs) have been widely applied in safety-critical tasks such as autopilot. However, when a perturbed input is fed into a DLS for inference, the DLS often has incorrect outputs (i.e., faults). DLS testing techniques (e.g., DeepXplore) detect such faults by generating perturbed inputs to explore data flows that induce faults. Since a DLS often has infinitely many data flows, existing techniques require developers to manually specify a set of activation values in a DLS’s neurons for exploring fault-inducing data flows. Unfortunately, recent studies show that such manual effort is tedious and can detect only a tiny proportion of fault-inducing data flows.In this paper, we present Themis, the first automatic DLS testing system, which attains strong fault detection capability by ensuring a full coverage of fault-inducing data flows at a high probability. Themis carries a new workflow for automatically and systematically revealing data flows whose internal neurons’ outputs vary substantially when the inputs are slightly perturbed, as these data flows are likely fault-inducing. We evaluated Themis on ten different DLSs and found that on average the number of faults detected by Themis was 3.78X more than four notable DLS testing techniques. By retraining all evaluated DLSs with the detected faults, Themis also increased (regained) these DLSs’ accuracies on average 14.7X higher than all baselines.

Large language models (LLMs) have shown remarkable progress in code generation, but their generated code often suffers from inefficiency, resulting in longer execution times and higher memory consumption. To address this issue, we propose EffiLearner, a self-optimization framework that utilizes execution overhead profiles to improve the efficiency of LLM-generated code. EffiLearner first generates code using an LLM, then executes it locally to capture execution time and memory usage profiles. These profiles are fed back to the LLM, which then revises the code to reduce overhead. To evaluate the effectiveness of EffiLearner, we conduct extensive experiments on EffiBench and two commonly used code generation benchmarks with 16 open-source and 6 closed-source models. Our evaluation results demonstrate that through iterative self-optimization, EffiLearner significantly enhances the efficiency of LLM-generated code. For example, the execution time (ET) of StarCoder2-15B for the EffiBench decreases from 0.93 (s) to 0.12 (s) which reduces 87.1\% execution time requirement compared with the initial code. The total memory usage (TMU) of StarCoder2-15B also decreases from 22.02 (Mb*s) to 2.03 (Mb*s), which decreases 90.8\% total memory consumption during the execution process.

Code generation models have increasingly become integral to aiding software development. Although current research has thoroughly examined the correctness of the code produced by code generation models, a vital aspect that plays a pivotal role in greencomputing and sustainability efforts — the efficiency of the generated code — has often been neglected. This paper presents Effibench, a benchmark with 1,000 efficiency-critical coding problems to assess the efficiency of code generated by code generation models. EffiBench contains a diverse set of LeetCode coding problems. Each problem is paired with an executable human-written canonical solution, which obtains the SOTA efficiency on the LeetCode solution leaderboard. With EffiBench, we empirically examine the ability of 42 large language models (35 open-source and 7 closed-source) to generate efficient code. Our evaluation results demonstrate that the efficiency of the code generated by LLMs is generally worse than the efficiency of human-written canonical solutions. For example, GPT-4 generated code has an average \textbf{3.12} times execution time that of the human-written canonical solutions. In the most extreme cases, the execution time and total memory usage of GPT-4 code are \textbf{13.89} and \textbf{43.92} times that of the canonical solutions. The source code of EffiBench is released on https://github.com/huangd1999/EffiBench. We also provide the LeaderBoard in https://huggingface.co/spaces/EffiBench/effibench-leaderboard.

Deep neural networks (DNNs) have been widely deployed in software to address various tasks (e.g., autonomous driving, medical diagnosis). However, they can also produce incorrect behaviors that result in financial losses and even threaten human safety. To reveal and repair incorrect behaviors in DNNs, developers often collect rich, unlabeled datasets from the natural world and label them to test DNN models. However, properly labeling a large number of datasets is a highly expensive and time-consuming task. To address the above-mentioned problem, we propose neuron sensitivity-guided test case selection (NSS), which can reduce the labeling time by selecting valuable test cases from unlabeled datasets. NSS leverages the information of the internal neuron induced by the test cases to select valuable test cases, which have high confidence in causing the model to behave incorrectly. We evaluated NSS with four widely used datasets and four well-designed DNN models compared to the state-of-the-art (SOTA) baseline methods. The results show that NSS performs well in assessing the probability of failure triggering in test cases and in the improvement capabilities of the model. Specifically, compared to the baseline approaches, NSS achieves a higher fault detection rate (e.g., when selecting 5% of the test cases from the unlabeled dataset in the MNIST and LeNet1 experiment, NSS can obtain an 81.8% fault detection rate, which is a 20% increase compared with SOTA baseline strategies).

The vulnerability of deep neural networks to adversarial samples has been a major impediment to their broad applications, despite their success in various fields. Recently, some works suggested that adversarially-trained models emphasize the importance of low-frequency information to achieve higher robustness. While several attempts have been made to leverage this frequency characteristic, they have all faced the issue that applying low-pass filters directly to input images leads to irreversible loss of discriminative information and poor generalizability to datasets with distinct frequency features. This paper presents a plug-and-play module called the Frequency Preference Control Module that adaptively reconfigures the low- and high-frequency components of intermediate feature representations, providing better utilization of frequency in robust learning. Empirical studies show that our proposed module can be easily incorporated into any adversarial training framework, further improving model robustness across different architectures and datasets. Additionally, experiments were conducted to examine how the frequency bias of robust models impacts the adversarial training process and its final robustness, revealing interesting insights.

The increasing computational complexity of DNNs achieved unprecedented successes in various areas such as machine vision and natural language processing (NLP), e.g., the recent advanced Transformer has billions of parameters. However, as large-scale DNNs significantly exceed GPU's physical memory limit, they cannot be trained by conventional methods such as data parallelism. Pipeline parallelism that partitions a large DNN into small subnets and trains them on different GPUs is a plausible solution. Unfortunately, the layer partitioning and memory management in existing pipeline parallel systems are fixed during training, making them easily impeded by out-of-memory errors and the GPU under-utilization. These drawbacks amplify when performing neural architecture search (NAS) such as the evolved Transformer, where different network architectures of Transformer needed to be trained repeatedly. vPipe is the first system that transparently provides dynamic layer partitioning and memory management for pipeline parallelism. vPipe has two unique contributions, including (1) an online algorithm for searching a near-optimal layer partitioning and memory management plan, and (2) a live layer migration protocol for re-balancing the layer distribution across a training pipeline. vPipe improved the training throughput of two notable baselines (Pipedream and GPipe) by 61.4-463.4 percent and 24.8-291.3 percent on various large DNNs and training settings.