Searcharxiv⌕ Search

arXiv subjects

Juntao Chen

Publications and source records attributed to Juntao Chen.

At least 19 recordsLinked to original sources
arXiv2026

Who Broke Me? Execution-Guided Repair of Behavioral Dependency Breaks

Dependency upgrades can break downstream projects without changing the library interface. Such behavioral breaking changes are difficult for developers to fix, because the failing test does not always point to the root API, the upstream API that causes the break. Existing LLM-based repair methods obtain evidence for the repair from compiler feedback or library documentation. However, a behavioral break produces no compiler feedback and is often undocumented. Agents that receive no upgrade evidence also usually do not retrieve upstream evidence themselves, and most of their failed repairs do not identify the root API. The library diff provides useful upstream evidence, but the root API must first be identified to select the relevant part of the diff. We present BBCFixer, a repair method that runs the failing test under the old and new library versions, ranks the calls whose return value differs to identify a candidate root API, and filters the library diff with that candidate. To evaluate repairs of behavioral breaks, we further introduce BBCBench, a benchmark of 100 behavioral dependency breaks in Python and JavaScript. We compare BBCFixer on BBCBench with two baselines: (1) Pure Agent, an agent that receives no upgrade evidence, and (2) BDUpdater, a method that mines library documentation. BBCFixer raises the average pass rate by up to 17% relative to Pure Agent and needs up to 27% fewer agent steps per successful repair. BBCFixer thus provides a way to select the upstream evidence behind a behavioral break, so that an LLM agent can repair behavioral breaking changes more effectively and efficiently.

cs.SE↗
arXiv2026

Ask the Expert: LLM-Guided Reinforcement Learning for Autonomous Cyber Defense

Policy-based reinforcement learning (RL) approaches have produced promising results for autonomous cyber defense; however, they are sample-inefficient in settings where defenders must respond under delayed, partial observations with actions from large action spaces. While large language models (LLMs) may reason semantically about security state space, high latency and trust assumptions prevent attractive in-line deployment models. We introduce Ask the Expert, a training-time guidance framework which first summarizes hard cyber-defense states, then intermittently queries an LLM for host-level defensive recommendations via a constrained action interface, and finally transforms those recommendations into tiered reward shaping for use with PPO. Because the LLM is discarded after training, deployment is a pure RL policy. Across TTCP CAGE CC1 and CC2 and both attacker types, this asymmetric design improves sample efficiency over PPO and outperforms the evaluated potential-based reward shaping (PBRS) baselines, while retaining the strongest terminal mean and requiring no LLM dependency at deployment time.

cs.CR↗
arXiv2026

Independent Multi-Agent Reinforcement Learning with Counterfactual Semantic-Social World Models

Fully decentralized multi-agent reinforcement learning (MARL), also referred to as independent learning, requires each agent to learn and act using only its local information and experience, without a centralized critic or inter-agent communication. Such a stringent information structure renders the conventional reward signal ambiguous. A poor return may result from an ineffective ego action, an incompatible teammate response, or an effective opponent response, yet scalar rewards alone do not reveal which explanation is responsible. We argue that agents can learn more effectively by prospectively comparing the consequences of candidate actions rather than diagnosing failures only from realized returns. We introduce CASTLE (Counterfactual Action-conditioned Semantic Tokens for Local Execution in Decentralized MARL), an offline-training, online-in-context guidance framework with two complementary world models. A Local Dynamics World Model, offline pre-trained over agents' local trajectories, summarizes the agent's local trajectory dynamics and partial observability, while a Semantic-Social World Model predicts compact short-horizon task and social consequences for each candidate ego action. The latter is trained from counterfactual simulator rollouts that expose plausible teammate and opponent responses to alternative actions taken from the same logged rollout state. During online learning and execution, both world models remain frozen and are queried by agents using only locally available information. Their prediction logits provide in-context guidance to an independent PPO policy. Across 30 matched seeds on Tag, Spread, and Adversary in the benchmark multi-particle environments, our proposed CASTLE achieves the highest mean final score among the evaluated methods, exceeding the strongest baseline on each task by 10.67, 6.46, and 0.33 normalized points, respectively.

cs.MA↗
arXiv2026

Decoder-Prior Poisoning in Quantum Error Correction: Attacks and PriorGuard Defense

Quantum error correction (QEC) protects quantum computations by repeatedly measuring stabilizer syndromes and using a classical decoder to infer corrections. Modern surface-code decoders are increasingly calibration-aware: they use recent device behavior to set priors such as matching-graph edge probabilities, and these priors directly shape the selected correction. We identify the prior-update path as an overlooked integrity-critical attack surface: a poisoned prior can change decoding decisions even when the syndrome stream, logical labels, and decoder implementation are unchanged. This makes prior poisoning different from ordinary syndrome anomalies and difficult for raw syndrome-anomaly tests to detect. We introduce PriorGuard, a lightweight semantic guard that checks proposed prior changes on high-influence detector-graph entries against private syndrome-derived evidence, accepts updates only when their log-odds movement is evidence-supported, and otherwise falls back to the last evidence-supported prior. We evaluate on Stim-generated rotated surface-code memory circuits decoded with PyMatching. Prior influence is highly concentrated, with the top 10 percent of entries carrying over 60 percent of positive logical-error influence; poisoning only 6.4 percent of decoder-prior entries can raise logical error rate (LER) by 3.35x. PriorGuard recovers most attack-induced loss with a low false-positive rate and modest update-boundary overhead of about 1.7--1.8 ms and less than 0.31 MB auxiliary memory per update.

quant-ph↗
arXiv2026

Hierarchical Agentic Incident Response with Digital-Twin-Validated Attack Inference

Network incident response remains slow and labor-intensive as the defender must infer multi-stage attacks from partial observations and translate recovery decisions into reliable system commands. Decision-theoretic planners provide principled optimization but typically rely on abstract states and predefined actions, while large language model (LLM) agents can reason over operational context but may hallucinate attacks and responses. Toward automating response planning, we present a hierarchical agentic response framework that integrates LLM-based attack inference, rollout planning, and digital-twin validation. A fine-tuned LLM infers the attack progression and affected hosts from security alerts and system measurements. An emulated network digital twin replays the inferred attack and returns discrepancies between predicted and observed effects to calibrate the inference. A separately fine-tuned planning agent uses the rollout planning method to prioritize affected components at the tactical layer. At the operational layer, the planning agent proposes high-level recovery actions, and an execution agent translates selected actions into recovery and verification commands that are validated in the digital twin. We evaluate the framework on a 33-component enterprise-network testbed under three multi-stage attack scenarios. The results show that our framework outperforms frontier-LLM baselines in recovery success rate by 18--31%.

cs.CR↗
arXiv2026

In-Context Reinforcement Learning via Communicative World Models

Reinforcement learning (RL) agents often struggle to generalize to new tasks and contexts without updating their parameters, mainly because their learned representations and policies are overfit to the specifics of their training environments. To boost agents' in-context RL (ICRL) ability, this work formulates ICRL as a two-agent emergent communication problem and introduces CORAL (Communicative Representation for Adaptive RL), a framework that learns a transferable communicative context by functionally separating latent representation learning from control. In CORAL, an Information Agent (IA) is pre-trained as a world model on a diverse distribution of tasks. Its objective is not direct return maximization, but world modeling and distilling its understanding into concise messages. The emergent communication protocol is shaped by a novel Causal Influence Loss, which measures the effect that the message has on the next action. During deployment, the previously trained IA serves as a fixed contextualizer for a new Control Agent (CA), which learns to solve tasks by interpreting the provided communicative context. Our experiments demonstrate that this approach enables the CA to achieve significant gains in sample efficiency and successfully perform zero-shot adaptation with the help of pre-trained IA in diverse online and offline environments, validating the efficacy of learning a transferable communicative representation.

cs.LG↗
arXiv2026

Breaking Changes in Software Ecosystems: A Systematic Literature Review

Modern software systems rely on dependency networks of reusable libraries, where breaking changes propagate and cause downstream consumers to fail. Despite growing research across ecosystems, no comprehensive synthesis exists. We conduct a systematic literature review of 97 primary studies, answering four research questions across five ecosystems: Maven/Java, npm/JavaScript, Python, Web APIs, and Linux distributions. The synthesis yields four results. First, a four-dimensional taxonomy along Nature, Detectability, Scope, and Visibility. Second, five reason categories and five impact dimensions, where maintenance and design improvements account for a larger share of breaking changes than new feature work. Third, 43 detection approaches that reach high accuracy on syntactic breaks but limited coverage on behavioral ones. Fourth, 66 strategies for communicating, preventing, and recovering from breaking changes, organized by the actor's role. Based on these findings, we identify three open challenges and three research opportunities. The challenges are behavioral break detection at scale, the failure of semantic versioning as a trust mechanism, and transitive dependency propagation under information asymmetry. The opportunities are LLM-augmented behavioral contract inference, ecosystem-level dependency graph intelligence, and domain-specific tooling for ML and data science.

cs.SE↗
arXiv2026

TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation

In this paper, we propose a sequential recommendation model that integrates Time-aware personalization, Multi-interest personalization, and Explanation personalization for Personalized Sequential Recommendation (TME-PSR). That is, we consider the differences across different users in temporal rhythm preference, multiple fine-grained latent interests, and the personalized semantic alignment between recommendations and explanations. Specifically, the proposed TME-PSR model employs a dual-view gated time encoder to capture personalized temporal rhythms, a lightweight multihead Linear Recurrent Unit architecture that enables fine-grained sub-interest modeling with improved efficiency, and a dynamic dual-branch mutual information weighting mechanism to achieve personalized alignment between recommendations and explanations. Extensive experiments on real-world datasets demonstrate that our method consistently improves recommendation accuracy and explanation quality, at a lower computational cost.

cs.IR↗
arXiv2026

SEA-Vision: A Multilingual Benchmark for Comprehensive Document and Scene Text Understanding in Southeast Asia

Multilingual document and scene text understanding plays an important role in applications such as search, finance, and public services. However, most existing benchmarks focus on high-resource languages and fail to evaluate models in realistic multilingual environments. In Southeast Asia, the diversity of languages, complex writing systems, and highly varied document types make this challenge even greater. We introduce SEA-Vision, a benchmark that jointly evaluates Document Parsing and Text-Centric Visual Question Answering (TEC-VQA) across 11 Southeast Asian languages. SEA-Vision contains 15,234 document parsing pages from nine representative document types, annotated with hierarchical page-, block-, and line-level labels. It also provides 7,496 TEC-VQA question-answer pairs that probe text recognition, numerical calculation, comparative analysis, logical reasoning, and spatial understanding. To make such multilingual, multi-task annotation feasible, we design a hybrid pipeline for Document Parsing and TEC-VQA. It combines automated filtering and scoring with MLLM-assisted labeling and lightweight native-speaker verification, greatly reducing manual labeling while maintaining high quality. We evaluate several leading multimodal models and observe pronounced performance degradation on low-resource Southeast Asian languages, highlighting substantial remaining gaps in multilingual document and scene text understanding. We believe SEA-Vision will help drive global progress in document and scene text understanding.

cs.CL↗
arXiv2026

Quantum-Audit: Evaluating the Reasoning Limits of LLMs on Quantum Computing

Language models have become practical tools for quantum computing education and research, from summarizing technical papers to explaining theoretical concepts and answering questions about recent developments in the field. While existing benchmarks evaluate quantum code generation and circuit design, their understanding of quantum computing concepts has not been systematically measured. Quantum-Audit addresses this gap with 2,700 questions covering core quantum computing topics. We evaluate 26 models from leading organizations. Our benchmark comprises 1,000 expert-written questions, 1,000 questions extracted from research papers using LLMs and validated by experts, plus an additional 700 questions including 350 open-ended questions and 350 questions with false premises to test whether models can correct erroneous assumptions. Human participants scored between 23% and 86%, with experts averaging 74%. Top-performing models exceeded the expert average, with Claude Opus 4.5 reaching 84% accuracy, though top models showed an average 12-point accuracy drop on expert-written questions compared to LLM-generated ones. Performance declined further on advanced topics, dropping to 73% on security questions. Additionally, models frequently accepted and reinforced false premises embedded in questions instead of identifying them, with accuracy below 66% on these critical reasoning tasks.

cs.CL↗
arXiv2026

Human-AI Co-Embodied Intelligence for Scientific Experimentation and Manufacturing

Scientific experimentation and manufacturing rely on prolonged protocol development and complex, multi-step implementation, which require continuous human expertise for precise execution and decision-making, limiting interpretability and scalability. Here, we introduce human-artificial intelligence (AI) co-embodied intelligence, a new form of physical AI that unites human researchers, agentic AI, and wearable hardware. In this paradigm, humans provide precise execution, while agentic AI contributes contextual reasoning, adaptive planning, and analysis. The wearable interface continuously captures experimentation and manufacturing, facilitating seamless communication between humans and AI. We instantiate this paradigm in a microfabrication cleanroom, leading to the agentic-physical experimentation (APEX) system which understands fabrication procedure with accuracy 51% higher than state-of-the-art multimodal large language models/vision language models (LLMs/VLMs), detects and corrects fabrication errors in real-time, and transfers procedural expertise to novice users. Critically, APEX system enables the co-development of fabrication protocols in cleanrooms, overcoming the incompatibility of elastomeric materials in standard microfabrication processes and enabling previously unattainable fabrication outcomes, as demonstrated by the wafer-scale realization of brain-level soft neural probe capable of single-unit-resolution neural recording. These results establish the human-AI co-embodied intelligence that extends agentic reasoning beyond computation into the physical domain, transforming scientific experimentation and manufacturing into autonomous, traceable, interpretable and scalable processes.

cs.AI↗
arXiv2026

Stackelberg Coupling of Online Representation Learning and Reinforcement Learning

Deep Q-learning jointly learns representations and values within monolithic networks, promising beneficial co-adaptation between features and value estimates. Although this architecture has attained substantial success, the coupling between representation and value learning creates instability as representations must constantly adapt to non-stationary value targets, while value estimates depend on these shifting representations. This is compounded by high variance in bootstrapped targets, which causes bias in value estimation in off-policy methods. We introduce Stackelberg Coupled Representation and Reinforcement Learning (SCORER), a framework for value-based RL that views representation and Q-learning as two strategic agents in a hierarchical game. SCORER models the Q-function as the leader, which commits to its strategy by updating less frequently, while the perception network (encoder) acts as the follower, adapting more frequently to learn representations that minimize Bellman error variance given the leader's committed strategy. Through this division of labor, the Q-function minimizes MSBE while perception minimizes its variance, thereby reducing bias accordingly, with asymmetric updates allowing stable co-adaptation, unlike simultaneous parameter updates in monolithic solutions. Our proposed SCORER framework leads to a bi-level optimization problem whose solution is approximated by a two-timescale algorithm that creates an asymmetric learning dynamic between the two players. Extensive experiments on DQN and its variants demonstrate that gains stem from algorithmic insight rather than model complexity.

cs.LG↗
arXiv2026

Differentiable Architecture Search for Adversarially Robust Quantum Computer Vision

Current quantum neural networks suffer from extreme sensitivity to both adversarial perturbations and hardware noise, creating a significant barrier to real-world deployment. Existing robustness techniques typically sacrifice clean accuracy or require prohibitive computational resources. We propose a hybrid quantum-classical Differentiable Quantum Architecture Search (DQAS) framework that addresses these limitations by jointly optimizing circuit structure and robustness through gradient-based methods. Our approach enhances traditional DQAS with a lightweight Classical Noise Layer applied before quantum processing, enabling simultaneous optimization of gate selection and noise parameters. This design preserves the quantum circuit's integrity while introducing trainable perturbations that enhance robustness without compromising standard performance. Experimental validation on MNIST, FashionMNIST, and CIFAR datasets shows consistent improvements in both clean and adversarial accuracy compared to existing quantum architecture search methods. Under various attack scenarios, including Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), Basic Iterative Method (BIM), and Momentum Iterative Method (MIM), and under realistic quantum noise conditions, our hybrid framework maintains superior performance. Testing on actual quantum hardware confirms the practical viability of discovered architectures. These results demonstrate that strategic classical preprocessing combined with differentiable quantum architecture optimization can significantly enhance quantum neural network robustness while maintaining computational efficiency.

quant-ph↗
arXiv2025

Hardware-aware and Resource-efficient Circuit Packing and Scheduling on Trapped-Ion Quantum Computers

The rapid expansion of quantum cloud services has led to long job queues due to single-tenant execution models that underutilize hardware resources. Quantum multi-programming (QMP) mitigates this by executing multiple circuits in parallel on a single device, but existing methods target superconducting systems with limited connectivity, high crosstalk, and lower gate fidelity. Trapped-ion architectures, with all-to-all connectivity, long coherence times, and high-fidelity mid-circuit measurement properties, presents itself as a more suitable platform for scalable QMP. We present CircPack, a hardware-aware circuit packing framework designed for modular trapped-ion devices based on the Quantum Charge-Coupled Device (QCCD) architecture. CircPack formulates static circuit scheduling as a two-dimensional packing problem with hardware-specific shuttling constraints. Compared to superconducting-based QMP approaches, CircPack achieves up to 70.72% better fidelity, 62.67% higher utilization, and 32.80% improved layer reduction. This framework is also capable of scalable, balanced scheduling across a cluster of independent QCCD modules, highlighting trapped-ion systems' potential in improving the throughput of quantum cloud computing in the near future.

quant-ph↗
arXiv2025

SCOUT: A Defense Against Data Poisoning Attacks in Fine-Tuned Language Models

Backdoor attacks create significant security threats to language models by embedding hidden triggers that manipulate model behavior during inference, presenting critical risks for AI systems deployed in healthcare and other sensitive domains. While existing defenses effectively counter obvious threats such as out-of-context trigger words and safety alignment violations, they fail against sophisticated attacks using contextually-appropriate triggers that blend seamlessly into natural language. This paper introduces three novel contextually-aware attack scenarios that exploit domain-specific knowledge and semantic plausibility: the ViralApp attack targeting social media addiction classification, the Fever attack manipulating medical diagnosis toward hypertension, and the Referral attack steering clinical recommendations. These attacks represent realistic threats where malicious actors exploit domain-specific vocabulary while maintaining semantic coherence, demonstrating how adversaries can weaponize contextual appropriateness to evade conventional detection methods. To counter both traditional and these sophisticated attacks, we present \textbf{SCOUT (Saliency-based Classification Of Untrusted Tokens)}, a novel defense framework that identifies backdoor triggers through token-level saliency analysis rather than traditional context-based detection methods. SCOUT constructs a saliency map by measuring how the removal of individual tokens affects the model's output logits for the target label, enabling detection of both conspicuous and subtle manipulation attempts. We evaluate SCOUT on established benchmark datasets (SST-2, IMDB, AG News) against conventional attacks (BadNet, AddSent, SynBkd, StyleBkd) and our novel attacks, demonstrating that SCOUT successfully detects these sophisticated threats while preserving accuracy on clean inputs.

cs.CR↗
arXiv2025

Analyzing and Optimizing the Distribution of Blood Lead Level Testing for Children in New York City: A Data-Driven Approach

This study investigates blood lead level (BLL) rates and testing among children under six years of age across the 42 neighborhoods in New York City from 2005 to 2021. Despite a citywide general decline in BLL rates, disparities at the neighborhood level persist and are not addressed in the official reports, highlighting the need for this comprehensive analysis. In this paper, we analyze the current BLL testing distribution and cluster the neighborhoods using a k-medoids clustering algorithm. We propose an optimized approach that improves resource allocation efficiency by accounting for case incidences and neighborhood risk profiles using a grid search algorithm. Our findings demonstrate statistically significant improvements in case detection and enhanced fairness by focusing on under-served and high-risk groups. Additionally, we propose actionable recommendations to raise awareness among parents, including outreach at local daycare centers and kindergartens, among other venues.

cs.CY↗
arXiv2025

Can LLMs Help Allocate Public Health Resources? A Case Study on Childhood Lead Testing

Public health agencies face critical challenges in identifying high-risk neighborhoods for childhood lead exposure with limited resources for outreach and intervention programs. To address this, we develop a Priority Score integrating untested children proportions, elevated blood lead prevalence, and public health coverage patterns to support optimized resource allocation decisions across 136 neighborhoods in Chicago, New York City, and Washington, D.C. We leverage these allocation tasks, which require integrating multiple vulnerability indicators and interpreting empirical evidence, to evaluate whether large language models (LLMs) with agentic reasoning and deep research capabilities can effectively allocate public health resources when presented with structured allocation scenarios. LLMs were tasked with distributing 1,000 test kits within each city based on neighborhood vulnerability indicators. Results reveal significant limitations: LLMs frequently overlooked neighborhoods with highest lead prevalence and largest proportions of untested children, such as West Englewood in Chicago, while allocating disproportionate resources to lower-priority areas like Hunts Point in New York City. Overall accuracy averaged 0.46, reaching a maximum of 0.66 with ChatGPT 5 Deep Research. Despite their marketed deep research capabilities, LLMs struggled with fundamental limitations in information retrieval and evidence-based reasoning, frequently citing outdated data and allowing non-empirical narratives about neighborhood conditions to override quantitative vulnerability indicators.

cs.CY↗
arXiv2025

Optimizing Inter-chip Coupler Link Placement for Modular and Chiplet Quantum Systems

Quantum computing offers unparalleled computational capabilities but faces significant challenges, including limited qubit counts, diverse hardware topologies, and dynamic noise and error rates, which hinder scalability and reliability. Distributed quantum computing, particularly chip-to-chip connections, has emerged as a solution by interconnecting multiple processors to collaboratively execute large circuits. While hardware advancements, such as IBM's Quantum Flamingo, focus on improving inter-chip fidelity, limited research addresses efficient circuit cutting and qubit mapping in distributed systems. This project introduces InterPlace, a self-adaptive, hardware-aware framework for chip-to-chip distributed quantum systems. InterPlace analyzes qubit noise and error rates to construct a virtual system topology, guiding circuit partitioning and distributed qubit mapping to minimize SWAP overhead and enhance fidelity. Implemented with IBM Qiskit and compared with the state-of-the-art, InterPlace achieves up to a 53.0\% improvement in fidelity and reduces the combination of on-chip SWAPs and inter-chip operations by as much as 33.3\%, demonstrating scalability and effectiveness in extensive evaluations on real quantum hardware topologies.

quant-ph↗