arXiv subjects

Bolun Zhang

Publications and source records attributed to Bolun Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Robust class-gated single-pixel diffractive optical neural network with random-aberration-aware training

Optical computing offers the theoretical potential for high-speed, energy-efficient inference, yet its practical deployment remains constrained by fundamental input-output bottlenecks, particularly the reliance on electronic sensors with limited frame rates and stringent alignment requirements between optical components. Here, we demonstrate an image-class-gated single-pixel DONN that overcomes these limitations by converting spatial complexity into a temporal intensity signature. Using a minimal architecture comprising a reconfigurable digital micromirror device and a single-pixel photodetector, we implement a virtual optical gate. The system time-multiplexes class-specific masks, causing the detector response to peak only when the mask index matches the input class. This allows the predicted label to be read out via peak timing rather than spatial localization, eliminating 2D sensor constraints. To bridge the persistent sim-to-real gap, we introduce a physics-aware training strategy using random-phase augmentation. This method renders the model intrinsically tolerant to phase aberrations and mechanical misalignments without requiring precise hardware modeling. Our prototype achieves 90.0%(MNIST) and 80.0% (Fashion-MNIST) accuracy at a readout rate of 5 kHz. By combining gigahertz-compatible single-pixel detection with robust and alignment-tolerant training, this work provides a scalable, hardware-efficient pathway toward real-time optical intelligent sensing.

physics.optics↗

arXiv2026

Rongzai agent: A Large Language Model-Based Autonomous Assistant for Rietveld Refinement of Neutron Diffraction Data

Neutron diffraction (ND) is an indispensable technique for determining atomic positions (especially light elements) and thus serves as a critical probe for revealing microscopic structures in materials science. However, traditional Rietveld refinement of ND data relies heavily on manual operation of specialized software, which is time-consuming, labor-intensive, and highly dependent on user expertise, severely hindering automated analysis. The automation of Rietveld refinement has long been a long-standing and challenging problem in crystallography. To address this challenge, this paper presents the Dr.Sai-Rongzai agent, an autonomous refinement assistant based on a large language model (LLM), a specialist knowledge base, and the GSAS-II refinement engine, achieving for the first time an intelligent refinement that integrates knowledge-driven decision-making. The agent accomplishes a fully automated workflow from natural language task parsing to autonomous decision-making, execution of refinement strategies, and report generation. Evaluation on five representative samples shows that the Rongzai agent achieves lower Rwp values than human specialists on three samples (2.88% vs. 4.42%, 5.06% vs. 5.40%, 7.60% vs. 9.00%), while on the other two samples its results are very close to those of the specialists. The agent is currently deployed at the China Spallation Neutron Source (CSNS) and is open for external user registration, providing an intelligent and user-friendly analytical tool for materials research. This work fully leverages the cutting-edge advantages of LLM, offers a new path to solve the long-standing problem of automated refinement, takes a key step toward intelligent and fully automated crystallographic analysis, and holds great potential to accelerate AI for Science discoveries in neutron-based materials characterization.

cond-mat.mtrl-sci↗

arXiv2026

HepScript: A Dual-Use DSL for Human-AI Collaborative Data Analysis Workflows in High-Energy Physics

The escalating data scale in High-Energy Physics (HEP) fuels a growing aspiration for higher analytical efficiency. While Large Language Models (LLMs) offer a path toward automation via agentic AI, they struggle with complex scientific workflows that require deep domain knowledge and are tightly coupled to experiment-specific codebases. To address this, we introduce a methodology centered on HepScript, a dual-use Domain-Specific Language (DSL) for HEP data analysis workflows. HepScript serves as a shared formal interface, abstracting HEP analysis logic into a constrained syntax that is both intuitive for human experts and reliably generable by AI agents. First developed for the Beijing Spectrometer III (BESIII) experiment, HepScript hides the complexity of the underlying software stack, translating high-level analysis intent into low-level, production-ready code. In our case studies, this abstraction reduces the required human-written code by 93\%. Crucially, HepScript's constrained grammar defines a tractable action space, enabling AI agents to autonomously generate executable specifications for core analysis stages directly from published literature with a 95\% success rate. Our work demonstrates a scalable pathway toward human-AI collaborative systems, where a formally specified DSL acts as an unambiguous translation layer between human expertise, AI automation, and production environment, rendering previously intractable automation problems solvable.

hep-ex↗

arXiv2026

Dr.Sai: An agentic AI for real-world physics analysis at BESIII

High Energy Physics (HEP) experiments like BESIII produce petabyte-scale data. Extracting physics results requires complex workflows (simulation, reconstruction, statistical analysis, etc.) that traditionally take experts months or years. Current manual methods are labor-intensive, prone to bias, and limit large-scale systematic scans. As data grows, this paradigm slows discovery. Large Language Models (LLMs) offer a solution. Their natural language understanding and code generation capabilities allow them to interpret scientific tasks and integrate with HEP tools (e.g., ROOT, BOSS) to act as an "AI partner" for autonomous analysis. We present Dr.Sai, an LLM-powered multi-agent system that translates natural language into rigorous physics workflows. As validation, Dr.Sai performed large-scale re-measurements of ten J/psi decay branching fractions - without manual coding. It successfully navigated the real BESIII computing environment and produced results matching established benchmarks. The article details Dr.Sai's architecture, the validation results, and performance evaluation. This work provides a blueprint for autonomous discovery, with relevance to other data-intensive fields like astronomy and genomics.

hep-ex↗

arXiv2025

Knowing Your Uncertainty -- On the application of LLM in social sciences

Large language models (LLMs) are rapidly being integrated into computational social science research, yet their blackboxed training and designed stochastic elements in inference pose unique challenges for scientific inquiry. This article argues that applying LLMs to social scientific tasks requires explicit assessment of uncertainty -- an expectation long established in both quantitative methodology in the social sciences and machine learning. We introduce a unified framework for evaluating LLM uncertainty based on Hill numbers, a family of diversity measures. By transforming existing uncertainty quantification (UQ) metrics into Hill numbers, the framework provides a common and intuitive scale for interpreting variation in LLM outputs while accommodating different notions of semantic similarity and different sensitivities to output distributions. We show how it might help the application of LLMs in social sciences through four empirical applications.

cs.CY↗

arXiv2025

Spectro-Polarimetric Holographic Multiplexing Metasurface with Super-High Capacity Empowered by Mechanical Rotation

Mechanically reconfigurable metasurfaces capable of translation, rotation, and permutation have attracted considerable attention for high-capacity optical information storage and full-color holographic displays, owing to their low-power and high functional scalability, despite the additional system-level complexity introduced by precision rotation stages. This study presents a differentiable inverse design framework for such metasurfaces, creating an accurate mapping between meta-atom geometries and their multi-channel optical responses across diverse optical dimensions. Using a deep neural network-driven, end-to-end optimization pipeline, the framework enables intelligent, iterative refinement of rotatable metasurface within constrained design space. Using this approach, we show high-fidelity holographic video display by rotating a single element in a cascaded metasurface doublet around the optical axis. The doublet enables pixel-resolved holographic imaging with 288 independent channels, and by switching input/output polarization states, the system demonstrates four distinct full-color dynamic holographic videos. This work establishes an alternative paradigm for optical parameter multiplexing and end-to-end inverse design in mechanically reconfigurable metasurfaces, suggesting applications in compact optical systems, dynamic holography, information processing, and optical computing.

physics.optics↗

arXiv2025

Automatically Benchmarking LLM Code Agents through Agent-Driven Annotation and Evaluation

Recent advances in code agents have enabled automated software development at the project level, supported by large language models (LLMs). However, existing benchmarks for code agent evaluation face two major limitations. First, creating high-quality project-level evaluation datasets requires extensive domain expertise, leading to prohibitive annotation costs and limited diversity. Second, while recent Agent-as-a-Judge paradigms address the rigidity of traditional unit tests by enabling flexible metrics, their reliance on In-Context Learning (ICL) with general LLMs often results in inaccurate assessments that misalign with human standards. To address these challenges, we propose an agent-driven benchmark construction pipeline that leverages human supervision to efficiently generate diverse project-level tasks. Based on this, we introduce PRDBench, comprising 50 real-world Python projects across 20 domains, each with structured Product Requirement Documents (PRDs) and comprehensive criteria. Furthermore, to overcome the inaccuracy of general LLM judges, we propose a highly reliable evaluation framework powered by a specialized, fine-tuned model. Based on Qwen3-Coder-30B, our dedicated PRDJudge achieves over 90% human alignment in fixed-interface scenarios. Extensive experiments demonstrate that our suite provides a scalable, robust, and highly accurate framework for assessing state-of-the-art code agents.

cs.SE↗

arXiv2025

CoreCodeBench: Decoupling Code Intelligence via Fine-Grained Repository-Level Tasks

The evaluation of Large Language Models (LLMs) for software engineering has shifted towards complex, repository-level tasks. However, existing benchmarks predominantly rely on coarse-grained pass rates that treat programming proficiency as a monolithic capability, obscuring specific cognitive bottlenecks. Furthermore, the static nature of these benchmarks renders them vulnerable to data contamination and performance saturation. To address these limitations, we introduce CoreCodeBench, a configurable repository-level benchmark designed to dissect coding capabilities through atomized tasks. Leveraging our automated framework, CorePipe, we extract and transform Python repositories into a comprehensive suite of tasks that isolate distinct cognitive demands within identical code contexts. Unlike static evaluations, CoreCodeBench supports controllable difficulty scaling to prevent saturation and ensures superior data quality. It achieves a 78.55% validity yield, significantly surpassing the 31.7% retention rate of SWE-bench-Verified. Extensive experiments with state-of-the-art LLMs reveal a significant capability misalignment, evidenced by distinct ranking shifts across cognitive dimensions. This indicates that coding proficiency is non-monolithic, as strength in one aspect does not necessarily translate to others. These findings underscore the necessity of our fine-grained taxonomy in diagnosing model deficiencies and offer a sustainable, rigorous framework for evolving code intelligence. The code for CorePipe is available at https://github.com/AGI-Eval-Official/CoreCodeBench, and the data for CoreCodeBench can be accessed at https://huggingface.co/collections/tubehhh/corecodebench-68256d2faabf4b1610a08caa.

cs.SE↗

arXiv2025

RoboBrain 2.0 Technical Report

We introduce RoboBrain 2.0, our latest generation of embodied vision-language foundation models, designed to unify perception, reasoning, and planning for complex embodied tasks in physical environments. It comes in two variants: a lightweight 7B model and a full-scale 32B model, featuring a heterogeneous architecture with a vision encoder and a language model. Despite its compact size, RoboBrain 2.0 achieves strong performance across a wide spectrum of embodied reasoning tasks. On both spatial and temporal benchmarks, the 32B variant achieves leading results, surpassing prior open-source and proprietary models. In particular, it supports key real-world embodied AI capabilities, including spatial understanding (e.g., affordance prediction, spatial referring, trajectory forecasting) and temporal decision-making (e.g., closed-loop interaction, multi-agent long-horizon planning, and scene graph updating). This report details the model architecture, data construction, multi-stage training strategies, infrastructure and practical applications. We hope RoboBrain 2.0 advances embodied AI research and serves as a practical step toward building generalist embodied agents. The code, checkpoint and benchmark are available at https://superrobobrain.github.io.

cs.RO↗

arXiv2025

Tady: A Neural Disassembler without Structural Constraint Violations

Disassembly is a crucial yet challenging step in binary analysis. While emerging neural disassemblers show promise for efficiency and accuracy, they frequently generate outputs violating fundamental structural constraints, which significantly compromise their practical usability. To address this critical problem, we regularize the disassembly solution space by formalizing and applying key structural constraints based on post-dominance relations. This approach systematically detects widespread errors in existing neural disassemblers' outputs. These errors often originate from models' limited context modeling and instruction-level decoding that neglect global structural integrity. We introduce Tady, a novel neural disassembler featuring an improved model architecture and a dedicated post-processing algorithm, specifically engineered to address these deficiencies. Comprehensive evaluations on diverse binaries demonstrate that Tady effectively eliminates structural constraint violations and functions with high efficiency, while maintaining instruction-level accuracy.

cs.CR↗

arXiv2025

Mono: Is Your "Clean" Vulnerability Dataset Really Solvable? Exposing and Trapping Undecidable Patches and Beyond

The quantity and quality of vulnerability datasets are essential for developing deep learning solutions to vulnerability-related tasks. Due to the limited availability of vulnerabilities, a common approach to building such datasets is analyzing security patches in source code. However, existing security patches often suffer from inaccurate labels, insufficient contextual information, and undecidable patches that fail to clearly represent the root causes of vulnerabilities or their fixes. These issues introduce noise into the dataset, which can mislead detection models and undermine their effectiveness. To address these issues, we present mono, a novel LLM-powered framework that simulates human experts' reasoning process to construct reliable vulnerability datasets. mono introduces three key components to improve security patch datasets: (i) semantic-aware patch classification for precise vulnerability labeling, (ii) iterative contextual analysis for comprehensive code understanding, and (iii) systematic root cause analysis to identify and filter undecidable patches. Our comprehensive evaluation on the MegaVul benchmark demonstrates that mono can correct 31.0% of labeling errors, recover 89% of inter-procedural vulnerabilities, and reveals that 16.7% of CVEs contain undecidable patches. Furthermore, mono's enriched context representation improves existing models' vulnerability detection accuracy by 15%. We open source the framework mono and the dataset MonoLens in https://github.com/vul337/mono.

cs.CR↗

arXiv2025

DecompileBench: A Comprehensive Benchmark for Evaluating Decompilers in Real-World Scenarios

Decompilers are fundamental tools for critical security tasks, from vulnerability discovery to malware analysis, yet their evaluation remains fragmented. Existing approaches primarily focus on syntactic correctness through synthetic micro-benchmarks or subjective human ratings, failing to address real-world requirements for semantic fidelity and analyst usability. We present DecompileBench, the first comprehensive framework that enables effective evaluation of decompilers in reverse engineering workflows through three key components: \textit{real-world function extraction} (comprising 23,400 functions from 130 real-world programs), \textit{runtime-aware validation}, and \textit{automated human-centric assessment} using LLM-as-Judge to quantify the effectiveness of decompilers in reverse engineering workflows. Through a systematic comparison between six industrial-strength decompilers and six recent LLM-powered approaches, we demonstrate that LLM-based methods surpass commercial tools in code understandability despite 52.2% lower functionality correctness. These findings highlight the potential of LLM-based approaches to transform human-centric reverse engineering. We open source \href{https://github.com/Jennieett/DecompileBench}{DecompileBench} to provide a framework to advance research on decompilers and assist security experts in making informed tool selections based on their specific requirements.

cs.SE↗

arXiv2025

Tinkering Against Scaling

The ascent of scaling in artificial intelligence research has revolutionized the field over the past decade, yet it presents significant challenges for academic researchers, particularly in computational social science and critical algorithm studies. The dominance of large language models, characterized by their extensive parameters and costly training processes, creates a disparity where only industry-affiliated researchers can access these resources. This imbalance restricts academic researchers from fully understanding their tools, leading to issues like reproducibility in computational social science and a reliance on black-box metaphors in critical studies. To address these challenges, we propose a "tinkering" approach that is inspired by existing works. This method involves engaging with smaller models or components that are manageable for ordinary researchers, fostering hands-on interaction with algorithms. We argue that tinkering is both a way of making and knowing for computational social science and a way of knowing for critical studies, and fundamentally, it is a way of caring that has broader implications for both fields.

cs.CY↗

arXiv2024

A Hybrid Quantum-Classical Autoencoder Framework for End-to-End Communication Systems

This paper investigates the application of quantum machine learning to End-to-End (E2E) communication systems in wireless fading scenarios. We introduce a novel hybrid quantum-classical autoencoder architecture that combines parameterized quantum circuits with classical deep neural networks (DNNs). Specifically, we propose a hybrid quantum-classical autoencoder (QAE) framework to optimize the E2E communication system. Our results demonstrate the feasibility of the proposed hybrid system, and reveal that it is the first work that can achieve comparable block error rate (BLER) performance to classical DNN-based and conventional channel coding schemes, while significantly reducing the number of trainable parameters. Additionally, the proposed QAE exhibits steady and superior BLER convergence over the classical autoencoder baseline.

cs.IT↗

arXiv2024

DDPG-E2E: A Novel Policy Gradient Approach for End-to-End Communication Systems

The End-to-end (E2E) learning-based approach has great potential to reshape the existing communication systems by replacing the transceivers with deep neural networks. To this end, the E2E learning approach needs to assume the availability of prior channel information to mathematically formulate a differentiable channel layer for the backpropagation (BP) of the error gradients, thereby jointly optimizing the transmitter and the receiver. However, accurate and instantaneous channel state information is hardly obtained in practical wireless communication scenarios. Moreover, the existing E2E learning-based solutions exhibit limited performance in data transmissions with large block lengths. In this article, these practical issues are addressed by our proposed deep deterministic policy gradient-based E2E communication system. In particular, the proposed solution utilizes a reward feedback mechanism to train both the transmitter and the receiver, which alleviates the information loss of error gradients during BP. In addition, a convolutional neural network (CNN)-based architecture is developed to mitigate the curse of dimensionality problem when transmitting messages with large block lengths. Extensive simulations then demonstrate that our proposed solution can not only jointly train the transmitter and the receiver simultaneously without requiring the prior channel knowledge but also can obtain significant performance improvement on block error rate compared to state-of-the-art solutions.

cs.NI↗

arXiv2024

Dynamic Spectrum Access for Ambient Backscatter Communication-assisted D2D Systems with Quantum Reinforcement Learning

Spectrum access is an essential problem in device-to-device (D2D) communications. However, with the recent growth in the number of mobile devices, the wireless spectrum is becoming scarce, resulting in low spectral efficiency for D2D communications. To address this problem, this paper aims to integrate the ambient backscatter communication technology into D2D devices to allow them to backscatter ambient RF signals to transmit their data when the shared spectrum is occupied by mobile users. To obtain the optimal spectrum access policy, i.e., stay idle or access the shared spectrum and perform active transmissions or backscattering ambient RF signals for transmissions, to maximize the average throughput for D2D users, deep reinforcement learning (DRL) can be adopted. However, DRL-based solutions may require long training time due to the curse of dimensionality issue as well as complex deep neural network architectures. For that, we develop a novel quantum reinforcement learning (RL) algorithm that can achieve a faster convergence rate with fewer training parameters compared to DRL thanks to the quantum superposition and quantum entanglement principles. Specifically, instead of using conventional deep neural networks, the proposed quantum RL algorithm uses a parametrized quantum circuit to approximate an optimal policy. Extensive simulations then demonstrate that the proposed solution not only can significantly improve the average throughput of D2D devices when the shared spectrum is busy but also can achieve much better performance in terms of convergence rate and learning complexity compared to existing DRL-based methods.

cs.NI↗

arXiv2024

Self-Alignment of Large Language Models via Monopolylogue-based Social Scene Simulation

Aligning large language models (LLMs) with human values is imperative to mitigate potential adverse effects resulting from their misuse. Drawing from the sociological insight that acknowledging all parties' concerns is a key factor in shaping human values, this paper proposes a novel direction to align LLMs by themselves: social scene simulation. To achieve this, we present MATRIX, a novel social scene simulator that emulates realistic scenes around a user's input query, enabling the LLM to take social consequences into account before responding. MATRIX serves as a virtual rehearsal space, akin to a Monopolylogue, where the LLM performs diverse roles related to the query and practice by itself. To inject this alignment, we fine-tune the LLM with MATRIX-simulated data, ensuring adherence to human values without compromising inference speed. We theoretically show that the LLM with MATRIX outperforms Constitutional AI under mild assumptions. Finally, extensive experiments validate that our method outperforms over 10 baselines across 4 benchmarks. As evidenced by 875 user ratings, our tuned 13B-size LLM exceeds GPT-4 in aligning with human values. See our project page at https://shuotang123.github.io/MATRIX.

cs.CL↗

arXiv2023

Deep Deterministic Policy Gradient for End-to-End Communication Systems without Prior Channel Knowledge

End-to-End (E2E) learning-based concept has been recently introduced to jointly optimize both the transmitter and the receiver in wireless communication systems. Unfortunately, this E2E learning architecture requires a prior differentiable channel model to jointly train the deep neural networks (DNNs) at the transceivers, which is hardly obtained in practice. This paper aims to solve this issue by developing a deep deterministic policy gradient (DDPG)-based framework. In particular, the proposed solution uses the loss value of the receiver DNN as the reward to train the transmitter DNN. The simulation results then show that our proposed solution can jointly train the transmitter and the receiver without requiring the prior channel model. In addition, we demonstrate that the proposed DDPG-based solution can achieve better detection performance compared to the state-of-the-art solutions.

cs.NI↗