arXiv subjects

Yu Kang

Publications and source records attributed to Yu Kang.

At least 37 records · Page 2Linked to original sources

arXiv2026

TestExplora: Benchmarking LLMs for Proactive Bug Discovery via Repository-Level Test Generation

Given that Large Language Models (LLMs) are increasingly applied to automate software development, comprehensive software assurance spans three distinct goals: regression prevention, reactive reproduction, and proactive discovery. Current evaluations systematically overlook the third goal. Specifically, they either treat existing code as ground truth (a compliance trap) for regression prevention, or depend on post-failure artifacts (e.g., issue reports) for bug reproduction-so they rarely surface defects before failures. To bridge this gap, we present TestExplora, a benchmark designed to evaluate LLMs as proactive testers within full-scale, realistic repository environments. TestExplora contains 2,389 tasks from 482 repositories and hides all defect-related signals. Models must proactively find bugs by comparing implementations against documentation-derived intent, using documentation as the oracle. Furthermore, to keep evaluation sustainable and reduce leakage, we propose continuous, time-aware data collection. Our evaluation reveals a significant capability gap: state-of-the-art models achieve a maximum Fail-to-Pass (F2P) rate of only 16.06%. Further analysis indicates that navigating complex cross-module interactions and leveraging agentic exploration are critical to advancing LLMs toward autonomous software quality assurance. Consistent with this, SWEAgent instantiated with GPT-5-mini achieves an F2P of 17.27% and an F2P@5 of 29.7%, highlighting the effectiveness and promise of agentic exploration in proactive bug discovery tasks.

cs.SE↗

arXiv2026

AntigenLM: Structure-Aware DNA Language Modeling for Influenza

Language models have advanced sequence analysis, yet DNA foundation models often lag behind task-specific methods for unclear reasons. We present AntigenLM, a generative DNA language model pretrained on influenza genomes with intact, aligned functional units. This structure-aware pretraining enables AntigenLM to capture evolutionary constraints and generalize across tasks. Fine-tuned on time-series hemagglutinin (HA) and neuraminidase (NA) sequences, AntigenLM accurately forecasts future antigenic variants across regions and subtypes, including those unseen during training, outperforming phylogenetic and evolution-based models. It also achieves near-perfect subtype classification. Ablation studies show that disrupting genomic structure through fragmentation or shuffling severely degrades performance, revealing the importance of preserving functional-unit integrity in DNA language modeling. AntigenLM thus provides both a powerful framework for antigen evolution prediction and a general principle for building biologically grounded DNA foundation models.

q-bio.GN↗

arXiv2026

Closing the Loop: Universal Repository Representation with RPG-Encoder

Current repository agents encounter a reasoning disconnect due to fragmented representations, as existing methods rely on isolated API documentation or dependency graphs that lack semantic depth. We consider repository comprehension and generation to be inverse processes within a unified cycle: generation expands intent into implementation, while comprehension compresses implementation back into intent. To address this, we propose RPG-Encoder, a framework that generalizes the Repository Planning Graph (RPG) from a static generative blueprint into a unified, high-fidelity representation. RPG-Encoder closes the reasoning loop through three mechanisms: (1) Encoding raw code into the RPG that combines lifted semantic features with code dependencies; (2) Evolving the topology incrementally to decouple maintenance costs from repository scale, reducing overhead by 95.7%; and (3) Operating as a unified interface for structure-aware navigation. In evaluations, RPG-Encoder establishes state-of-the-art localization performance on SWE-bench Verified with 93.7% Acc@5 and exceeds the best baseline by over 10% in localization accuracy on SWE-bench Live Lite. These results highlight our superior fine-grained precision in complex codebases. Furthermore, it achieves 98.5% reconstruction coverage on RepoCraft, confirming RPG's high-fidelity capacity to mirror the original codebase and closing the loop between intent and implementation.

cs.CL↗

arXiv2026

DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems

Large language model (LLM)-based multi-agent systems are challenging to debug because failures often arise from long, branching interaction traces. The prevailing practice is to leverage LLMs for log-based failure localization, attributing errors to a specific agent and step. However, this paradigm has two key limitations: (i) log-only debugging lacks validation, producing untested hypotheses, and (ii) single-step or single-agent attribution is often ill-posed, as we find that multiple distinct interventions can independently repair the failed task. To address the first limitation, we introduce DoVer, an intervention-driven debugging framework, which augments hypothesis generation with active verification through targeted interventions (e.g., editing messages, altering plans). For the second limitation, rather than evaluating on attribution accuracy, we focus on measuring whether the system resolves the failure or makes quantifiable progress toward task success, reflecting a more outcome-oriented view of debugging. Within the Magnetic-One agent framework, on the datasets derived from GAIA and AssistantBench, DoVer flips 18-28% of failed trials into successes, achieves up to 16% milestone progress, and validates or refutes 30-60% of failure hypotheses. DoVer also performs effectively on a different dataset (GSMPlus) and agent framework (AG2), where it recovers 49% of failed trials. These results highlight intervention as a practical mechanism for improving reliability in agentic systems and open opportunities for more robust, scalable debugging methods for LLM-based multi-agent systems. Project website and code will be available at https://aka.ms/DoVer.

cs.AI↗

arXiv2026

DA-SPS: A Dual-stage Network based on Singular Spectrum Analysis, Patching-strategy and Spearman-correlation for Multivariate Time-series Prediction

Multivariate time-series forecasting, as a typical problem in the field of time series prediction, has a wide range of applications in weather forecasting, traffic flow prediction, and other scenarios. However, existing works do not effectively consider the impact of extraneous variables on the prediction of the target variable. On the other hand, they fail to fully extract complex sequence information based on various time patterns of the sequences. To address these drawbacks, we propose a DA-SPS model, which adopts different modules for feature extraction based on the information characteristics of different variables. DA-SPS mainly consists of two stages: the target variable processing stage (TVPS) and the extraneous variables processing stage (EVPS). In TVPS, the model first uses Singular Spectrum Analysis (SSA) to process the target variable sequence and then uses Long Short-Term Memory (LSTM) and P-Conv-LSTM which deploys a patching strategy to extract features from trend and seasonality components, respectively. In EVPS, the model filters extraneous variables that have a strong correlation with the target variate by using Spearman correlation analysis and further analyses them using the L-Attention module which consists of LSTM and attention mechanism. Finally, the results obtained by TVPS and EVPS are combined through weighted summation and linear mapping to produce the final prediction. The results on four public datasets demonstrate that the DA-SPS model outperforms existing state-of-the-art methods. Additionally, its performance in real-world scenarios is further validated using a private dataset collected by ourselves, which contains the test items' information on laptop motherboards.

cs.LG↗

arXiv2026

IIB-LPO: Latent Policy Optimization via Iterative Information Bottleneck

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) for Large Language Model (LLM) reasoning have been hindered by a persistent challenge: exploration collapse. The semantic homogeneity of random rollouts often traps models in narrow, over-optimized behaviors. While existing methods leverage policy entropy to encourage exploration, they face inherent limitations. Global entropy regularization is susceptible to reward hacking, which can induce meaningless verbosity, whereas local token-selective updates struggle with the strong inductive bias of pre-trained models. To address this, we propose Latent Policy Optimization via Iterative Information Bottleneck (IIB-LPO), a novel approach that shifts exploration from statistical perturbation of token distributions to topological branching of reasoning trajectories. IIB-LPO triggers latent branching at high-entropy states to diversify reasoning paths and employs the Information Bottleneck principle both as a trajectory filter and a self-reward mechanism, ensuring concise and informative exploration. Empirical results across four mathematical reasoning benchmarks demonstrate that IIB-LPO achieves state-of-the-art performance, surpassing prior methods by margins of up to 5.3% in accuracy and 7.4% in diversity metrics.

cs.LG↗

arXiv2025

Zoomer: Adaptive Image Focus Optimization for Black-box MLLM

Multimodal large language models (MLLMs) such as GPT-4o, Gemini Pro, and Claude 3.5 have enabled unified reasoning over text and visual inputs, yet they often hallucinate in real world scenarios especially when small objects or fine spatial context are involved. We pinpoint two core causes of this failure: the absence of region-adaptive attention and inflexible token budgets that force uniform downsampling, leading to critical information loss. To overcome these limitations, we introduce Zoomer, a visual prompting framework that delivers token-efficient, detail-preserving image representations for black-box MLLMs. Zoomer integrates (1) a prompt-aware emphasis module to highlight semantically relevant regions, (2) a spatial-preserving orchestration schema to maintain object relationships, and (3) a budget-aware strategy to adaptively allocate tokens between global context and local details. Extensive experiments on nine benchmarks and three commercial MLLMs demonstrate that Zoomer boosts accuracy by up to 27% while cutting image token usage by up to 67%. Our approach establishes a principled methodology for robust, resource-aware multimodal understanding in settings where model internals are inaccessible.

cs.CV↗

arXiv2025

C3RL: Rethinking the Combination of Channel-independence and Channel-mixing from Representation Learning

Multivariate time series forecasting has drawn increasing attention due to its practical importance. Existing approaches typically adopt either channel-mixing (CM) or channel-independence (CI) strategies. CM strategy can capture inter-variable dependencies but fails to discern variable-specific temporal patterns. CI strategy improves this aspect but fails to fully exploit cross-variable dependencies like CM. Hybrid strategies based on feature fusion offer limited generalization and interpretability. To address these issues, we propose C3RL, a novel representation learning framework that jointly models both CM and CI strategies. Motivated by contrastive learning in computer vision, C3RL treats the inputs of the two strategies as transposed views and builds a siamese network architecture: one strategy serves as the backbone, while the other complements it. By jointly optimizing contrastive and prediction losses with adaptive weighting, C3RL balances representation and forecasting performance. Extensive experiments on seven models show that C3RL boosts the best-case performance rate to 81.4% for models based on CI strategy and to 76.3% for models based on CM strategy, demonstrating strong generalization and effectiveness.

cs.LG↗

arXiv2025

Continuous Benchmark Generation for Evaluating Enterprise-scale LLM Agents

The rapid adoption of AI agents across domains has made systematic evaluation crucial for ensuring their usefulness and successful production deployment. Evaluation of AI agents typically involves using a fixed set of benchmarks and computing multiple evaluation metrics for the agent. While sufficient for simple coding tasks, these benchmarks fall short for enterprise-scale agents, where services and requirements evolve continuously and ground-truth examples are sparse. We propose a process of benchmark generation that helps evolve the benchmarks as the requirements change and perform robust evaluation of evolving AI agents. We instantiate this approach for a case study of service migration from one deployment platform to another at a large public enterprise. Our approach relies on semi-structured documents where developers express the high-level intent, and uses state-of-the-art LLMs to generate benchmarks from just a small number of such documents. Overall, this process results in a maintainable evaluation framework, enabling rapid feedback on agent performance and facilitating targeted improvements.

cs.SE↗

arXiv2025

GUI-360$^\circ$: A Comprehensive Dataset and Benchmark for Computer-Using Agents

We introduce GUI-360$^\circ$, a large-scale, comprehensive dataset and benchmark suite designed to advance computer-using agents (CUAs). CUAs present unique challenges and is constrained by three persistent gaps: a scarcity of real-world CUA tasks, the lack of automated collection-and-annotation pipelines for multi-modal trajectories, and the absence of a unified benchmark that jointly evaluates GUI grounding, screen parsing, and action prediction. GUI-360$^\circ$ addresses these gaps with an LLM-augmented, largely automated pipeline for query sourcing, environment-template construction, task instantiation, batched execution, and LLM-driven quality filtering. The released corpus contains over 1.2M executed action steps across thousands of trajectories in popular Windows office applications, and includes full-resolution screenshots, accessibility metadata when available, instantiated goals, intermediate reasoning traces, and both successful and failed action trajectories. The dataset supports three canonical tasks, GUI grounding, screen parsing, and action prediction, and a hybrid GUI+API action space that reflects modern agent designs. Benchmarking state-of-the-art vision--language models on GUI-360$^\circ$ reveals substantial out-of-the-box shortcomings in grounding and action prediction; supervised fine-tuning and reinforcement learning yield significant gains but do not close the gap to human-level reliability. We release GUI-360$^\circ$ and accompanying code to facilitate reproducible research and accelerate progress on robust desktop CUAs. The full dataset has been made public on https://huggingface.co/datasets/vyokky/GUI-360.

cs.AI↗

arXiv2025

Towards Large-Scale In-Context Reinforcement Learning by Meta-Training in Randomized Worlds

In-Context Reinforcement Learning (ICRL) enables agents to learn automatically and on-the-fly from their interactive experiences. However, a major challenge in scaling up ICRL is the lack of scalable task collections. To address this, we propose the procedurally generated tabular Markov Decision Processes, named AnyMDP. Through a carefully designed randomization process, AnyMDP is capable of generating high-quality tasks on a large scale while maintaining relatively low structural biases. To facilitate efficient meta-training at scale, we further introduce decoupled policy distillation and induce prior information in the ICRL framework. Our results demonstrate that, with a sufficiently large scale of AnyMDP tasks, the proposed model can generalize to tasks that were not considered in the training set through versatile in-context learning paradigms. The scalable task set provided by AnyMDP also enables a more thorough empirical investigation of the relationship between data distribution and ICRL performance. We further show that the generalization of ICRL potentially comes at the cost of increased task diversity and longer adaptation periods. This finding carries critical implications for scaling robust ICRL capabilities, highlighting the necessity of diverse and extensive task design, and prioritizing asymptotic performance over few-shot adaptation.

cs.LG↗

arXiv2025

ExeCoder: Empowering Large Language Models with Executability Representation for Code Translation

Code translation is a crucial activity in the software development and maintenance process, and researchers have recently begun to focus on using pre-trained large language models (LLMs) for code translation. However, existing LLMs only learn the contextual semantics of code during pre-training, neglecting executability information closely related to the execution state of the code, which results in unguaranteed code executability and unreliable automated code translation. To address this issue, we propose ExeCoder, an LLM specifically designed for code translation, aimed at utilizing executability representations such as functional semantics, syntax structures, and variable dependencies to enhance the capabilities of LLMs in code translation. To evaluate the effectiveness of ExeCoder, we manually enhanced the widely used benchmark TransCoder-test, resulting in a benchmark called TransCoder-test-X that serves LLMs. Evaluation of TransCoder-test-X indicates that ExeCoder achieves state-of-the-art performance in code translation, surpassing existing open-source code LLMs by over 10.88% to 38.78% and over 27.44% to 42.97% on two metrics, and even outperforms the renowned closed-source LLM GPT-4o. Code is available at https://aka.ms/execoder

cs.SE↗

arXiv2025

Scale-Disentangled spatiotemporal Modeling for Long-term Traffic Emission Forecasting

Long-term traffic emission forecasting is crucial for the comprehensive management of urban air pollution. Traditional forecasting methods typically construct spatiotemporal graph models by mining spatiotemporal dependencies to predict emissions. However, due to the multi-scale entanglement of traffic emissions across time and space, these spatiotemporal graph modeling method tend to suffer from cascading error amplification during long-term inference. To address this issue, we propose a Scale-Disentangled Spatio-Temporal Modeling (SDSTM) framework for long-term traffic emission forecasting. It leverages the predictability differences across multiple scales to decompose and fuse features at different scales, while constraining them to remain independent yet complementary. Specifically, the model first introduces a dual-stream feature decomposition strategy based on the Koopman lifting operator. It lifts the scale-coupled spatiotemporal dynamical system into an infinite-dimensional linear space via Koopman operator, and delineates the predictability boundary using gated wavelet decomposition. Then a novel fusion mechanism is constructed, incorporating a dual-stream independence constraint based on cross-term loss to dynamically refine the dual-stream prediction results, suppress mutual interference, and enhance the accuracy of long-term traffic emission prediction. Extensive experiments conducted on a road-level traffic emission dataset within Xi'an's Second Ring Road demonstrate that the proposed model achieves state-of-the-art performance.

cs.LG↗

arXiv2025

Set-Valued Transformer Network for High-Emission Mobile Source Identification

Identifying high-emission vehicles is a crucial step in regulating urban pollution levels and formulating traffic emission reduction strategies. However, in practical monitoring data, the proportion of high-emission state data is significantly lower compared to normal emission states. This characteristic long-tailed distribution severely impedes the extraction of discriminative features for emission state identification during data mining. Furthermore, the highly nonlinear nature of vehicle emission states and the lack of relevant prior knowledge also pose significant challenges to the construction of identification models.To address the aforementioned issues, we propose a Set-Valued Transformer Network (SVTN) to achieve comprehensive learning of discriminative features from high-emission samples, thereby enhancing detection accuracy. Specifically, this model first employs the transformer to measure the temporal similarity of micro-trip condition variations, thus constructing a mapping rule that projects the original high-dimensional emission data into a low-dimensional feature space. Next, a set-valued identification algorithm is used to probabilistically model the relationship between the generated feature vectors and their labels, providing an accurate metric criterion for the classification algorithm. To validate the effectiveness of our proposed approach, we conducted extensive experiments on the diesel vehicle monitoring data of Hefei city in 2020. The results demonstrate that our method achieves a 9.5\% reduction in the missed detection rate for high-emission vehicles compared to the transformer-based baseline, highlighting its superior capability in accurately identifying high-emission mobile pollution sources.

cs.LG↗

arXiv2025

WarriorMath: Enhancing the Mathematical Ability of Large Language Models with a Defect-aware Framework

Large Language Models (LLMs) excel in solving mathematical problems, yet their performance is often limited by the availability of high-quality, diverse training data. Existing methods focus on augmenting datasets through rephrasing or difficulty progression but overlook the specific failure modes of LLMs. This results in synthetic questions that the model can already solve, providing minimal performance gains. To address this, we propose WarriorMath, a defect-aware framework for mathematical problem solving that integrates both targeted data synthesis and progressive training. In the synthesis stage, we employ multiple expert LLMs in a collaborative process to generate, critique, and refine problems. Questions that base LLMs fail to solve are identified and iteratively improved through expert-level feedback, producing high-quality, defect-aware training data. In the training stage, we introduce a progressive learning framework that iteratively fine-tunes the model using increasingly challenging data tailored to its weaknesses. Experiments on six mathematical benchmarks show that WarriorMath outperforms strong baselines by 12.57% on average, setting a new state-of-the-art. Our results demonstrate the effectiveness of a defect-aware, multi-expert framework for improving mathematical ability.

cs.CL↗

arXiv2025

DC-Mamber: A Dual Channel Prediction Model based on Mamba and Linear Transformer for Multivariate Time Series Forecasting

In multivariate time series forecasting (MTSF), existing strategies for processing sequences are typically categorized as channel-independent and channel-mixing. The former treats all temporal information of each variable as a token, focusing on capturing local temporal features of individual variables, while the latter constructs a token from the multivariate information at each time step, emphasizing the modeling of global temporal dependencies. Current mainstream models are mostly based on Transformer and the emerging Mamba. Transformers excel at modeling global dependencies through self-attention mechanisms but exhibit limited sensitivity to local temporal patterns and suffer from quadratic computational complexity, restricting their efficiency in long-sequence processing. In contrast, Mamba, based on state space models (SSMs), achieves linear complexity and efficient long-range modeling but struggles to aggregate global contextual information in parallel. To overcome the limitations of both models, we propose DC-Mamber, a dual-channel forecasting model based on Mamba and linear Transformer for time series forecasting. Specifically, the Mamba-based channel employs a channel-independent strategy to extract intra-variable features, while the Transformer-based channel adopts a channel-mixing strategy to model cross-timestep global dependencies. DC-Mamber first maps the raw input into two distinct feature representations via separate embedding layers. These representations are then processed by a variable encoder (built on Mamba) and a temporal encoder (built on linear Transformer), respectively. Finally, a fusion layer integrates the dual-channel features for prediction. Extensive experiments on eight public datasets confirm DC-Mamber's superior accuracy over existing models.

cs.AI↗

arXiv2025

A Scalable and Quantum-Accurate Foundation Model for Biomolecular Force Field via Linearly Tensorized Quadrangle Attention

Accurate atomistic biomolecular simulations are vital for disease mechanism understanding, drug discovery, and biomaterial design, but existing simulation methods exhibit significant limitations. Classical force fields are efficient but lack accuracy for transition states and fine conformational details critical in many chemical and biological processes. Quantum Mechanics (QM) methods are highly accurate but computationally infeasible for large-scale or long-time simulations. AI-based force fields (AIFFs) aim to achieve QM-level accuracy with efficiency but struggle to balance many-body modeling complexity, accuracy, and speed, often constrained by limited training data and insufficient validation for generalizability. To overcome these challenges, we introduce LiTEN, a novel equivariant neural network with Tensorized Quadrangle Attention (TQA). TQA efficiently models three- and four-body interactions with linear complexity by reparameterizing high-order tensor features via vector operations, avoiding costly spherical harmonics. Building on LiTEN, LiTEN-FF is a robust AIFF foundation model, pre-trained on the extensive nablaDFT dataset for broad chemical generalization and fine-tuned on SPICE for accurate solvated system simulations. LiTEN achieves state-of-the-art (SOTA) performance across most evaluation subsets of rMD17, MD22, and Chignolin, outperforming leading models such as MACE, NequIP, and EquiFormer. LiTEN-FF enables the most comprehensive suite of downstream biomolecular modeling tasks to date, including QM-level conformer searches, geometry optimization, and free energy surface construction, while offering 10x faster inference than MACE-OFF for large biomolecules (~1000 atoms). In summary, we present a physically grounded, highly efficient framework that advances complex biomolecular modeling, providing a versatile foundation for drug discovery and related applications.

physics.chem-ph↗

arXiv2025

API Agents vs. GUI Agents: Divergence and Convergence

Large language models (LLMs) have evolved beyond simple text generation to power software agents that directly translate natural language commands into tangible actions. While API-based LLM agents initially rose to prominence for their robust automation capabilities and seamless integration with programmatic endpoints, recent progress in multimodal LLM research has enabled GUI-based LLM agents that interact with graphical user interfaces in a human-like manner. Although these two paradigms share the goal of enabling LLM-driven task automation, they diverge significantly in architectural complexity, development workflows, and user interaction models. This paper presents the first comprehensive comparative study of API-based and GUI-based LLM agents, systematically analyzing their divergence and potential convergence. We examine key dimensions and highlight scenarios in which hybrid approaches can harness their complementary strengths. By proposing clear decision criteria and illustrating practical use cases, we aim to guide practitioners and researchers in selecting, combining, or transitioning between these paradigms. Ultimately, we indicate that continuing innovations in LLM-based automation are poised to blur the lines between API- and GUI-driven agents, paving the way for more flexible, adaptive solutions in a wide range of real-world applications.

cs.AI↗