Searcharxiv⌕ Search

arXiv subjects

Ziru Chen

Publications and source records attributed to Ziru Chen.

At least 19 recordsLinked to original sources

Mobile Edge Deployment and Resource Management for Maritime Wireless Networks

Mobile Edge Computing (MEC) has been envisioned as one of the key technologies for supplying computation and storage resources in Internet of Vessels (IoV) networks. Due to its flexible deployment, low cost and agile maneuverability, Unmanned Surface Vehicle (USV) has emerged as a promising solution, to provide communication and computation services for maritime users. In this paper, we study mobile edge deployment and resource management for MEC-assisted maritime wireless networks where USVs with diverse computation resources are deployed to provide edge computing services that complement the cloud-based services. To this end, we formulate an optimization problem to minimize the expected response time by jointly optimizing the deployment of mobile USVs and computation offloading decisions. To solve the mixed-integer nonlinear program problem, we propose a Dual-Layer Reinforcement Learning (DLRL) framework to attain a near-optimal solution. Specifically, a Deep Deterministic Policy Gradient (DDPG) algorithm is designed to obtain the best USV deployment in the outer layer learning, and a Q-learning algorithm is designed to determine the best computation offloading decisions in the inner layer learning. Numerical results demonstrate that the proposed solution outperforms some literature algorithms by effectively handling both continuous and discrete variables.

cs.NI↗

COBALT: Contextual Bandit Learning with Offline Trajectories for Multi-Turn Code Generation

Training large language models (LLMs) for multi-turn code generation typically relies on online reinforcement learning (RL), in which models learn from repeatedly generated online rollouts. Although effective, online RL can be costly and unstable. In this paper, we present contextual bandit learning with offline trajectories (COBALT), a simple method that reuses offline multi-turn trajectories to efficiently train LLMs for multi-turn code generation. Building on the observation that multi-turn code generation is a one-step recoverable Markov decision process, COBALT first collects code generation trajectories using a reference LLM and divides them into partial trajectories. Then, during online bandit learning, the LLM is trained to complete each partial trajectory through single-step code generation. On LiveCodeBench, COBALT substantially improves R1-Distill 8B and Qwen3 8B by up to 9.0 and 6.2 absolute Pass@1 points and matches the performance of two multi-turn online RL baselines using GRPO and VeRPO. We also perform a perturbation-based analysis of in-context reward hacking and show that Cobalt can improve robustness to misleading execution feedback. Overall, our results demonstrate that Cobalt is a promising solution for multi-turn code generation. Our code and data are available at https://github.com/OSU-NLP-Group/cobalt.

cs.LG↗

FREA: A Multi-Source Expert Benchmark for Reaction Feasibility Verification

As generative models and AI agents propose chemical reactions at a scale beyond expert review, feasibility verifiers decide which proposals enter synthesis planning. But do their decisions agree with chemists across different kinds of candidates? We introduce FREA, a benchmark of 751 reactions labeled by expert chemists under an explicit feasibility criterion, drawn from retrosynthesis model proposals, zero-yield experimental records, edits by large language models (LLMs), and five negative candidate generation methods. Our evaluation finds that no verifier leads across all sources: LLMs given only the criterion are competitive with dedicated verifiers, while forward models perform best on retrosynthesis proposals but reject most feasible edits of recorded reactions at the evaluated operating points. Looking beyond aggregate scores, both forward models perform below chance when separating infeasible alternative disconnections from feasible generated candidates. To study whether negative supervision addresses these weaknesses, we also release a corpus of over 14 million recorded reactions and generated negative candidates. In matched training comparisons, adding a mixture of generated negatives to forward training raises mean AUROC across sources, but these gains do not extend to retrosynthesis proposals. Varying the generation method further shows that the largest gain on generated candidates coincides with worse proposal screening. These findings motivate evaluating verifiers against experts across sources and designing negatives for transfer to model proposals.

cs.AI↗

AutoSDT: Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists

Despite long-standing efforts in accelerating scientific discovery with AI, building AI co-scientists remains challenging due to limited high-quality data for training and evaluation. To tackle this data scarcity issue, we present AutoSDT, an automatic pipeline that collects high-quality coding tasks in real-world data-driven discovery workflows. AutoSDT leverages the coding capabilities and parametric knowledge of LLMs to search for diverse sources, select ecologically valid tasks, and synthesize accurate task instructions and code solutions. Using our pipeline, we construct AutoSDT-5K, a dataset of 5,404 coding tasks for data-driven discovery that covers four scientific disciplines and 756 unique Python packages. To the best of our knowledge, AutoSDT-5K is the only automatically collected and the largest open dataset for data-driven scientific discovery. Expert feedback on a subset of 256 tasks shows the effectiveness of AutoSDT: 93% of the collected tasks are ecologically valid, and 92.2% of the synthesized programs are functionally correct. Trained on AutoSDT-5K, the Qwen2.5-Coder-Instruct LLM series, dubbed AutoSDT-Coder, show substantial improvement on two challenging data-driven discovery benchmarks, ScienceAgentBench and DiscoveryBench. Most notably, AutoSDT-Coder-32B reaches the same level of performance as GPT-4o on ScienceAgentBench with a success rate of 7.8%, doubling the performance of its base model. On DiscoveryBench, it lifts the hypothesis matching score to 8.1, bringing a 17.4% relative improvement and closing the gap between open-weight models and GPT-4o.

cs.LG↗

D3-Gym: Constructing Real-World Verifiable Environments for Data-Driven Discovery

Despite recent progress in language models and agents for scientific data-driven discovery, advancing their capabilities is held back by the absence of verifiable environments representing real-world scientific tasks. To fill this gap, we introduce D3-Gym, the first automatically constructed dataset with verifiable environments for scientific Data-Driven Discovery. D3-Gym comprises 565 tasks from 239 real scientific repositories across four disciplines, each with a natural language instruction, an executable environment with pre-installed dependencies, dataset previews, a reference solution, and an automatically synthesized evaluation script. Our evaluation scripts achieve 87.5% agreement with human-annotated gold standards and strong alignment in domain-specific evaluation logic. Training on trajectories sampled from D3-Gym yields consistent gains across Qwen3 models on ScienceAgentBench, boosting Qwen3-32B by 7.8 absolute points and shrinking the gap with strong proprietary models. We further illustrate, through case studies, how D3-Gym environments can serve as a testbed for studying agentic optimization loops such as Autoresearch on real scientific workflows. We open-source D3-Gym, its creation workflow, sampled trajectories, and training scripts at https://github.com/OSU-NLP-Group/D3-Gym.

cs.AI↗

Near-Field Dual-UPA Communications: A Generalized Geometric Approach

This paper investigates a near-field (NF) multiple-input multiple-output (MIMO) communication system equipped with dual uniform planar arrays (UPAs). We first develop a generalized geometric model to calculate the 3D distance between arbitrary antenna elements across the transmitter and receiver panels. Leveraging the distance analysis, we derive a closed-form near-field to far-field (NF-FF) boundary for dual-UPA configurations. By exploiting the geometric structure of the UPAs, we further decompose the near-field channel matrix into a Kronecker-product of two lower-dimensional matrices. This decomposition enables a low-complexity NF beamforming design for achievable-rate maximization. Numerical results validate the analysis and demonstrate that the conventional Rayleigh distance is a special case of the generalized model. Furthermore, the proposed beamforming design achieves near-optimal rate performance while significantly reducing the computational complexity compared to state-of-the-art NF beamforming methods.

eess.SP↗

Joint Beamforming Design and Port Selection in Fluid Antenna-Assisted Multi-Cell Networks: A Personalized Federated Learning Approach

This paper investigates joint beamforming and port selection in multi-cell fluid antenna-assisted (FAS) networks. In such networks, active beamforming and discrete FA port selection are coupled through intra-cell and inter-cell interference and are jointly optimized to maximize the weighted sum-rate (WSR). We develop a federated representation learning (FedRep) framework with a position-aware dual-branch deep neural network (PA-DNN). The PA-DNN uses channel state information and port positional encoding as inputs, and jointly outputs beamforming vectors and port selections through two task-specific branches. To support decentralized training across heterogeneous cells, the FedRep framework shares global beamforming-related parameters among base stations while keeping port-selection parameters local for cell-specific adaptation. Simulation results show that the proposed scheme achieves a higher weighted sum-rate than conventional FL and port-selection benchmark schemes.

cs.IT↗

LC4-DViT: Land-cover Creation for Land-cover Classification with Deformable Vision Transformer

Land-cover underpins ecosystem services, hydrologic regulation, disaster-risk reduction, and evidence-based land planning; timely, accurate land-cover maps are therefore critical for environmental stewardship. Remote sensing-based land-cover classification offers a scalable route to such maps but is hindered by scarce and imbalanced annotations and by geometric distortions in high-resolution scenes. We propose LC4-DViT (Land-cover Creation for Land-cover Classification with Deformable Vision Transformer), a framework that combines generative data creation with a deformation-aware Vision Transformer. A text-guided diffusion pipeline uses GPT-4o-generated scene descriptions and super-resolved exemplars to synthesize class-balanced, high-fidelity training images, while DViT couples a DCNv4 deformable convolutional backbone with a Vision Transformer encoder to jointly capture fine-scale geometry and global context. On eight classes from the Aerial Image Dataset (AID)-Beach, Bridge, Desert, Forest, Mountain, Pond, Port, and River-DViT achieves 0.9572 overall accuracy, 0.9576 macro F1-score, and 0.9510 Cohen' s Kappa, improving over a vanilla ViT baseline (0.9274 OA, 0.9300 macro F1, 0.9169 Kappa) and outperforming ResNet50, MobileNetV2, and FlashInternImage. Cross-dataset experiments on a three-class SIRI-WHU subset (Harbor, Pond, River) yield 0.9333 overall accuracy, 0.9316 macro F1, and 0.8989 Kappa, indicating good transferability. An LLM-based judge using GPT-4o to score Grad-CAM heatmaps further shows that DViT' s attention aligns best with hydrologically meaningful structures. These results suggest that description-driven generative augmentation combined with deformation-aware transformers is a promising approach for high-resolution land-cover mapping.

cs.CV↗

Hierarchical Reinforcement Learning for Next Generation of Multi-AP Coordinated Spatial Reuse

In next generation of Wi-Fi networks Multiple Access Point Coordination (MAPC) is poised to significantly enhance the network performance by enabling a set of Access Points (APs) to coordinate with each other through advanced coordinating schemes so that to reduce inter-AP contention and congestion. This paper focuses on defining a framework to facilitate the coordination across multi-APs when these employ Coordinated Spatial Reuse (C-SR). In this case, the coordinating APs may need to reciprocally adjust their scheduling strategy, power control and link adaptation to meet specific Quality of Service (QoS) requirements, which by using classical approaches leads to high overhead due to negotiations needed across APs, and requires complex solutions in order to properly optimize the network across all the parameters in play. In this matter, a two layer Multi-Armed Bandit (MAB) algorithm has been proposed to optimize such a network while preserving the fair use of resources across all nodes. The validity of this holistic approach is confirmed by system level simulations, which show that the proposed algorithm not only improves the network in terms of sum-throughput, but also allows to enhance fairness, making this a robust solution for next-generation of Wi-Fi networks.

cs.NI↗

Agent Data Protocol: Unifying Datasets for Diverse, Effective Fine-tuning of LLM Agents

Public research results on large-scale supervised finetuning of AI agents remain relatively rare, since the collection of agent training data presents unique challenges. In this work, we argue that the bottleneck is not a lack of underlying data sources, but that a large variety of data is fragmented across heterogeneous formats, tools, and interfaces. To this end, we introduce the agent data protocol (ADP), a light-weight representation language that serves as an "interlingua" between agent datasets in diverse formats and unified agent training pipelines downstream. The design of ADP is expressive enough to capture a large variety of tasks, including API/tool use, browsing, coding, software engineering, and general agentic workflows, while remaining simple to parse and train on without engineering at a per-dataset level. In experiments, we unified a broad collection of 13 existing agent training datasets into ADP format, and converted the standardized ADP data into training-ready formats for multiple agent frameworks. We performed SFT on these data, and demonstrated an average performance gain of ~20% over corresponding base models, and delivers state-of-the-art or near-SOTA performance on standard coding, browsing, tool use, and research benchmarks, without domain-specific tuning. All code and data are released publicly, in the hope that ADP could help lower the barrier to standardized, scalable, and reproducible agent training.

cs.CL↗

MVT: Mask-Grounded Vision-Language Models for Taxonomy-Aligned Land-Cover Tagging

Land-cover understanding in remote sensing increasingly demands class-agnostic systems that generalize across datasets while remaining spatially precise and interpretable. We study a geometry-first discovery-and-interpretation setting under domain shift, where candidate regions are delineated class-agnostically and supervision avoids lexical class names via anonymized identifiers. Complementary to open-set recognition and open-world learning, we focus on coupling class-agnostic mask evidence with taxonomy-grounded scene interpretation, rather than unknown rejection or continual class expansion. We propose MVT, a three-stage framework that (i) extracts boundary-faithful region masks using SAM2 with domain adaptation, (ii) performs mask-grounded semantic tagging and scene description generation via dual-step LoRA fine-tuning of multimodal LLMs, and (iii) evaluates outputs with LLM-as-judge scoring calibrated by stratified expert ratings. On cross-dataset segmentation transfer (train on OpenEarthMap, evaluate on LoveDA), domain-adapted SAM2 improves mask quality; meanwhile, dual-step MLLM fine-tuning yields more accurate taxonomy-aligned tags and more informative mask-grounded scene descriptions.

cs.CV↗

Holistic Agent Leaderboard: The Missing Infrastructure for AI Agent Evaluation

AI agents have been developed for complex real-world tasks from coding to customer service. But AI agent evaluations suffer from many challenges that undermine our understanding of how well agents really work. We introduce the Holistic Agent Leaderboard (HAL) to address these challenges. We make three main contributions. First, we provide a standardized evaluation harness that orchestrates parallel evaluations across hundreds of VMs, reducing evaluation time from weeks to hours while eliminating common implementation bugs. Second, we conduct three-dimensional analysis spanning models, scaffolds, and benchmarks. We validate the harness by conducting 21,730 agent rollouts across 9 models and 9 benchmarks in coding, web navigation, science, and customer service with a total cost of about $40,000. Our analysis reveals surprising insights, such as higher reasoning effort reducing accuracy in the majority of runs. Third, we use LLM-aided log inspection to uncover previously unreported behaviors, such as searching for the benchmark on HuggingFace instead of solving a task, or misusing credit cards in flight booking tasks. We share all agent logs, comprising 2.5B tokens of language model calls, to incentivize further research into agent behavior. By standardizing how the field evaluates agents and addressing common pitfalls in agent evaluation, we hope to shift the focus from agents that ace benchmarks to agents that work reliably in the real world.

cs.AI↗

Large Language Models Achieve Gold Medal Performance at the International Olympiad on Astronomy & Astrophysics (IOAA)

While task-specific demonstrations show early success in applying large language models (LLMs) to automate some astronomical research tasks, they only provide incomplete views of all necessary capabilities in solving astronomy problems, calling for more thorough understanding of LLMs' strengths and limitations. So far, existing benchmarks and evaluations focus on simple question-answering that primarily tests astronomical knowledge and fails to evaluate the complex reasoning required for real-world research in the discipline. Here, we address this gap by systematically benchmarking five state-of-the-art LLMs on the International Olympiad on Astronomy and Astrophysics (IOAA) exams, which are designed to examine deep conceptual understanding, multi-step derivations, and multimodal analysis. With average scores of 85.6% and 84.2%, Gemini 2.5 Pro and GPT-5 (the two top-performing models) not only achieve gold medal level performance but also rank in the top two among ~200-300 participants in all four IOAA theory exams evaluated (2022-2025). In comparison, results on the data analysis exams show more divergence. GPT-5 still excels in the exams with an 88.5% average score, ranking top 10 among the participants in the four most recent IOAAs, while other models' performances drop to 48-76%. Furthermore, our in-depth error analysis underscores conceptual reasoning, geometric reasoning, and spatial visualization (52-79% accuracy) as consistent weaknesses among all LLMs. Hence, although LLMs approach peak human performance in theory exams, critical gaps must be addressed before they can serve as autonomous research agents in astronomy.

astro-ph.IM↗

GeoAnalystBench: A GeoAI benchmark for assessing large language models for spatial analysis workflow and code generation

Recent advances in large language models (LLMs) have fueled growing interest in automating geospatial analysis and GIS workflows, yet their actual capabilities remain uncertain. In this work, we call for rigorous evaluation of LLMs on well-defined geoprocessing tasks before making claims about full GIS automation. To this end, we present GeoAnalystBench, a benchmark of 50 Python-based tasks derived from real-world geospatial problems and carefully validated by GIS experts. Each task is paired with a minimum deliverable product, and evaluation covers workflow validity, structural alignment, semantic similarity, and code quality (CodeBLEU). Using this benchmark, we assess both proprietary and open source models. Results reveal a clear gap: proprietary models such as ChatGPT-4o-mini achieve high validity 95% and stronger code alignment (CodeBLEU 0.39), while smaller open source models like DeepSeek-R1-7B often generate incomplete or inconsistent workflows (48.5% validity, 0.272 CodeBLEU). Tasks requiring deeper spatial reasoning, such as spatial relationship detection or optimal site selection, remain the most challenging across all models. These findings demonstrate both the promise and limitations of current LLMs in GIS automation and provide a reproducible framework to advance GeoAI research with human-in-the-loop support.

cs.SE↗

ChemToolAgent: The Impact of Tools on Language Agents for Chemistry Problem Solving

To enhance large language models (LLMs) for chemistry problem solving, several LLM-based agents augmented with tools have been proposed, such as ChemCrow and Coscientist. However, their evaluations are narrow in scope, leaving a large gap in understanding the benefits of tools across diverse chemistry tasks. To bridge this gap, we develop ChemToolAgent, an enhanced chemistry agent over ChemCrow, and conduct a comprehensive evaluation of its performance on both specialized chemistry tasks and general chemistry questions. Surprisingly, ChemToolAgent does not consistently outperform its base LLMs without tools. Our error analysis with a chemistry expert suggests that: For specialized chemistry tasks, such as synthesis prediction, we should augment agents with specialized tools; however, for general chemistry questions like those in exams, agents' ability to reason correctly with chemistry knowledge matters more, and tool augmentation does not always help.

cs.AI↗

ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery

The advancements of large language models (LLMs) have piqued growing interest in developing LLM-based language agents to automate scientific discovery end-to-end, which has sparked both excitement and skepticism about their true capabilities. In this work, we call for rigorous assessment of agents on individual tasks in a scientific workflow before making bold claims on end-to-end automation. To this end, we present ScienceAgentBench, a new benchmark for evaluating language agents for data-driven scientific discovery. To ensure the scientific authenticity and real-world relevance of our benchmark, we extract 102 tasks from 44 peer-reviewed publications in four disciplines and engage nine subject matter experts to validate them. We unify the target output for every task to a self-contained Python program file and employ an array of evaluation metrics to examine the generated programs, execution results, and costs. Each task goes through multiple rounds of manual validation by annotators and subject matter experts to ensure its annotation quality and scientific plausibility. We also propose two effective strategies to mitigate data contamination concerns. Using ScienceAgentBench, we evaluate five open-weight and proprietary LLMs, each with three frameworks: direct prompting, OpenHands CodeAct, and self-debug. Given three attempts for each task, the best-performing agent can only solve 32.4% of the tasks independently and 34.3% with expert-provided knowledge. In addition, we evaluate OpenAI o1-preview with direct prompting and self-debug, which can boost the performance to 42.2%, demonstrating the effectiveness of increasing inference-time compute but with more than 10 times the cost of other LLMs. Still, our results underscore the limitations of current language agents in generating code for data-driven discovery, let alone end-to-end automation for scientific research.

cs.CL↗

Cost Optimization for Serverless Edge Computing with Budget Constraints using Deep Reinforcement Learning

Serverless computing adopts a pay-as-you-go billing model where applications are executed in stateless and shortlived containers triggered by events, resulting in a reduction of monetary costs and resource utilization. However, existing platforms do not provide an upper bound for the billing model which makes the overall cost unpredictable, precluding many organizations from managing their budgets. Due to the diverse ranges of serverless functions and the heterogeneous capacity of edge devices, it is challenging to receive near-optimal solutions for deployment cost in a polynomial time. In this paper, we investigated the function scheduling problem with a budget constraint for serverless computing in wireless networks. Users and IoT devices are sending requests to edge nodes, improving the latency perceived by users. We propose two online scheduling algorithms based on reinforcement learning, incorporating several important characteristics of serverless functions. Via extensive simulations, we justify the superiority of the proposed algorithm by comparing with an ILP solver (Midaco). Our results indicate that the proposed algorithms efficiently approximate the results of Midaco within a factor of 1.03 while our decision-making time is 5 orders of magnitude less than that of Midaco.

cs.NI↗

Context-aware Container Orchestration in Serverless Edge Computing

Adopting serverless computing to edge networks benefits end-users from the pay-as-you-use billing model and flexible scaling of applications. This paradigm extends the boundaries of edge computing and remarkably improves the quality of services. However, due to the heterogeneous nature of computing and bandwidth resources in edge networks, it is challenging to dynamically allocate different resources while adapting to the burstiness and high concurrency in serverless workloads. This article focuses on serverless function provisioning in edge networks to optimize end-to-end latency, where the challenge lies in jointly allocating wireless bandwidth and computing resources among heterogeneous computing nodes. To address this challenge, We devised a context-aware learning framework that adaptively orchestrates a wide spectrum of resources and jointly considers them to avoid resource fragmentation. Extensive simulation results justified that the proposed algorithm reduces over 95% of converge time while the end-to-end delay is comparable to the state of the art.

cs.NI↗