arXiv subjects

Bo Shen

Publications and source records attributed to Bo Shen.

At least 55 records · Page 3Linked to original sources

arXiv2024

Deep Computer Vision for Solar Physics Big Data: Opportunities and Challenges

With recent missions such as advanced space-based observatories like the Solar Dynamics Observatory (SDO) and Parker Solar Probe, and ground-based telescopes like the Daniel K. Inouye Solar Telescope (DKIST), the volume, velocity, and variety of data have made solar physics enter a transformative era as solar physics big data (SPBD). With the recent advancement of deep computer vision, there are new opportunities in SPBD for tackling problems that were previously unsolvable. However, there are new challenges arising due to the inherent characteristics of SPBD and deep computer vision models. This vision paper presents an overview of the different types of SPBD, explores new opportunities in applying deep computer vision to SPBD, highlights the unique challenges, and outlines several potential future research directions.

cs.CV↗

arXiv2024

SWE-bench-java: A GitHub Issue Resolving Benchmark for Java

GitHub issue resolving is a critical task in software engineering, recently gaining significant attention in both industry and academia. Within this task, SWE-bench has been released to evaluate issue resolving capabilities of large language models (LLMs), but has so far only focused on Python version. However, supporting more programming languages is also important, as there is a strong demand in industry. As a first step toward multilingual support, we have developed a Java version of SWE-bench, called SWE-bench-java. We have publicly released the dataset, along with the corresponding Docker-based evaluation environment and leaderboard, which will be continuously maintained and updated in the coming months. To verify the reliability of SWE-bench-java, we implement a classic method SWE-agent and test several powerful LLMs on it. As is well known, developing a high-quality multi-lingual benchmark is time-consuming and labor-intensive, so we welcome contributions through pull requests or collaboration to accelerate its iteration and refinement, paving the way for fully automated programming.

cs.SE↗

arXiv2024

On Model Compression for Neural Networks: Framework, Algorithm, and Convergence Guarantee

Model compression is a crucial part of deploying neural networks (NNs), especially when the memory and storage of computing devices are limited in many applications. This paper focuses on two model compression techniques: low-rank approximation and weight pruning in neural networks, which are very popular nowadays. However, training NN with low-rank approximation and weight pruning always suffers significant accuracy loss and convergence issues. In this paper, a holistic framework is proposed for model compression from a novel perspective of nonconvex optimization by designing an appropriate objective function. Then, we introduce NN-BCD, a block coordinate descent (BCD) algorithm to solve the nonconvex optimization. One advantage of our algorithm is that an efficient iteration scheme can be derived with closed-form, which is gradient-free. Therefore, our algorithm will not suffer from vanishing/exploding gradient problems. Furthermore, with the Kurdyka-Łojasiewicz (KŁ) property of our objective function, we show that our algorithm globally converges to a critical point at the rate of O(1/k), where k denotes the number of iterations. Lastly, extensive experiments with tensor train decomposition and weight pruning demonstrate the efficiency and superior performance of the proposed framework. Our code implementation is available at https://github.com/ChenyangLi-97/NN-BCD

cs.LG↗

arXiv2024

Towards Uncovering How Large Language Model Works: An Explainability Perspective

Large language models (LLMs) have led to breakthroughs in language tasks, yet the internal mechanisms that enable their remarkable generalization and reasoning abilities remain opaque. This lack of transparency presents challenges such as hallucinations, toxicity, and misalignment with human values, hindering the safe and beneficial deployment of LLMs. This paper aims to uncover the mechanisms underlying LLM functionality through the lens of explainability. First, we review how knowledge is architecturally composed within LLMs and encoded in their internal parameters via mechanistic interpretability techniques. Then, we summarize how knowledge is embedded in LLM representations by leveraging probing techniques and representation engineering. Additionally, we investigate the training dynamics through a mechanistic perspective to explain phenomena such as grokking and memorization. Lastly, we explore how the insights gained from these explanations can enhance LLM performance through model editing, improve efficiency through pruning, and better align with human values.

cs.CL↗

arXiv2024

CodeS: Natural Language to Code Repository via Multi-Layer Sketch

The impressive performance of large language models (LLMs) on code-related tasks has shown the potential of fully automated software development. In light of this, we introduce a new software engineering task, namely Natural Language to code Repository (NL2Repo). This task aims to generate an entire code repository from its natural language requirements. To address this task, we propose a simple yet effective framework CodeS, which decomposes NL2Repo into multiple sub-tasks by a multi-layer sketch. Specifically, CodeS includes three modules: RepoSketcher, FileSketcher, and SketchFiller. RepoSketcher first generates a repository's directory structure for given requirements; FileSketcher then generates a file sketch for each file in the generated structure; SketchFiller finally fills in the details for each function in the generated file sketch. To rigorously assess CodeS on the NL2Repo task, we carry out evaluations through both automated benchmarking and manual feedback analysis. For benchmark-based evaluation, we craft a repository-oriented benchmark, SketchEval, and design an evaluation metric, SketchBLEU. For feedback-based evaluation, we develop a VSCode plugin for CodeS and engage 30 participants in conducting empirical studies. Extensive experiments prove the effectiveness and practicality of CodeS on the NL2Repo task.

cs.CL↗

arXiv2024

Single-Shot Single-Beam Coherent Raman Scattering Thermometry Based on Air Lasing

Thermometric techniques with high accuracy, fast response speed and ease of implementation are desirable for the study of dynamic combustion environments, transient reacting flows, and non-equilibrium plasmas. Herein, single-shot single-beam coherent Raman scattering (SS-CRS) thermometry is developed, for the first time to our knowledge, by using air lasing as a probe. It's proved that the air-lasing-assisted CRS signal has a high signal-to-noise ratio enabling single-shot measurements at a 1 kHz repetition rate. The SS-CRS thermometry consistently exhibits precision better than 2% at different temperatures, but the inaccuracy grows with the increase in temperature. The high detection precision, 1 kHz acquisition rate and easy-to-implement single-beam scheme are achieved thanks to the unique temporal, spectral and spatial characteristics of air lasing. This work opens a novel avenue for high-speed CRS thermometry, holding tremendous potential for fast diagnostics of transient reacting flows and plasmas.

physics.optics↗

arXiv2024

CoderEval: A Benchmark of Pragmatic Code Generation with Generative Pre-trained Models

Code generation models based on the pre-training and fine-tuning paradigm have been increasingly attempted by both academia and industry, resulting in well-known industrial models such as Codex, CodeGen, and PanGu-Coder. To evaluate the effectiveness of these models, multiple existing benchmarks are proposed, including only cases of generating a standalone function, i.e., a function that may invoke or access only built-in functions and standard libraries. However, non-standalone functions, which typically are not included in the existing benchmarks, constitute more than 70% of the functions in popular open-source projects, and evaluating models' effectiveness on standalone functions cannot reflect these models' effectiveness on pragmatic code generation scenarios. To help bridge the preceding gap, in this paper, we propose a benchmark named CoderEval, consisting of 230 Python and 230 Java code generation tasks carefully curated from popular real-world open-source projects and a self-contained execution platform to automatically assess the functional correctness of generated code. CoderEval supports code generation tasks from six levels of context dependency, where context refers to code elements such as types, APIs, variables, and consts defined outside the function under generation but within the dependent third-party libraries, current class, file, or project. CoderEval can be used to evaluate the effectiveness of models in generating code beyond only standalone functions. By evaluating three code generation models on CoderEval, we find that the effectiveness of these models in generating standalone functions is substantially higher than that in generating non-standalone functions. Our analysis highlights the current progress and pinpoints future directions to further improve a model's effectiveness by leveraging contextual information for pragmatic code generation.

cs.SE↗

arXiv2023

Linear RNNs Provably Learn Linear Dynamic Systems

We study the learning ability of linear recurrent neural networks with Gradient Descent. We prove the first theoretical guarantee on linear RNNs to learn any stable linear dynamic system using any a large type of loss functions. For an arbitrary stable linear system with a parameter $ρ_C$ related to the transition matrix $C$, we show that despite the non-convexity of the parameter optimization loss if the width of the RNN is large enough (and the required width in hidden layers does not rely on the length of the input sequence), a linear RNN can provably learn any stable linear dynamic system with the sample and time complexity polynomial in $\frac{1}{1-ρ_C}$. Our results provide the first theoretical guarantee to learn a linear RNN and demonstrate how can the recurrent structure help to learn a dynamic system.

cs.LG↗

arXiv2023

Can Programming Languages Boost Each Other via Instruction Tuning?

When human programmers have mastered a programming language, it would be easier when they learn a new programming language. In this report, we focus on exploring whether programming languages can boost each other during the instruction fine-tuning phase of code large language models. We conduct extensive experiments of 8 popular programming languages (Python, JavaScript, TypeScript, C, C++, Java, Go, HTML) on StarCoder. Results demonstrate that programming languages can significantly improve each other. For example, CodeM-Python 15B trained on Python is able to increase Java by an absolute 17.95% pass@1 on HumanEval-X. More surprisingly, we found that CodeM-HTML 7B trained on the HTML corpus can improve Java by an absolute 15.24% pass@1. Our training data is released at https://github.com/NL2Code/CodeM.

cs.CL↗

arXiv2023

Effect of Grain Coalescence on Dislocation and Stress Evolution of GaN Films Grown on Nanoscale Patterned Sapphire Substrates

Two types of nucleation layers (NLs), including in-situ low-temperature grown GaN (LT-GaN) and ex-situ sputtered physical vapor deposition AlN (PVD-AlN), are applied on cone-shaped nanoscale patterned sapphire substrate (NPSS). The initial growth process of GaN on these two NLs is comparably investigated by a series of growth interruptions. The coalescence process of GaN grains is modulated by adjusting the three-dimensional (3D) temperatures. The results indicate that higher 3D temperatures reduce the edge dislocation density while increasing the residual compressive stress in GaN films. Compared to the LT-GaN NLs, the PVD-AlN NLs effectively resist Ostwald ripening and facilitate the uniform growth of GaN grains on NPSS. Furthermore, GaN films grown on NPSS with PVD-AlN NLs exhibit a reduction of over 50% in both screw and edge dislocation densities compared to those grown on LT-GaN NLs. Additionally, PVD-AlN NLs result in an increase of about 0.5 GPa in the residual compressive stress observed in GaN films.

cond-mat.mtrl-sci↗

arXiv2023

PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback

Large Language Models for Code (Code LLM) are flourishing. New and powerful models are released on a weekly basis, demonstrating remarkable performance on the code generation task. Various approaches have been proposed to boost the code generation performance of pre-trained Code LLMs, such as supervised fine-tuning, instruction tuning, reinforcement learning, etc. In this paper, we propose a novel RRTF (Rank Responses to align Test&Teacher Feedback) framework, which can effectively and efficiently boost pre-trained large language models for code generation. Under this framework, we present PanGu-Coder2, which achieves 62.20% pass@1 on the OpenAI HumanEval benchmark. Furthermore, through an extensive evaluation on CoderEval and LeetCode benchmarks, we show that PanGu-Coder2 consistently outperforms all previous Code LLMs.

cs.CL↗

arXiv2022

Anomaly Detection in Additive Manufacturing Processes using Supervised Classification with Imbalanced Sensor Data based on Generative Adversarial Network

Supervised classification methods have been widely utilized for the quality assurance of the advanced manufacturing process, such as additive manufacturing (AM) for anomaly (defects) detection. However, since abnormal states (with defects) occur much less frequently than normal ones (without defects) in a manufacturing process, the number of sensor data samples collected from a normal state is usually much more than that from an abnormal state. This issue causes imbalanced training data for classification analysis, thus deteriorating the performance of detecting abnormal states in the process. It is beneficial to generate effective artificial sample data for the abnormal states to make a more balanced training set. To achieve this goal, this paper proposes a novel data augmentation method based on a generative adversarial network (GAN) using additive manufacturing process image sensor data. The novelty of our approach is that a standard GAN and classifier are jointly optimized with techniques to stabilize the learning process of standard GAN. The diverse and high-quality generated samples provide balanced training data to the classifier. The iterative optimization between GAN and classifier provides the high-performance classifier. The effectiveness of the proposed method is validated by both open-source data and real-world case studies in polymer and metal AM processes.

cs.LG↗

arXiv2022

A Novel Sparse Bayesian Learning and Its Application to Fault Diagnosis for Multistation Assembly Systems

This paper addresses the problem of fault diagnosis in multistation assembly systems. Fault diagnosis is to identify process faults that cause the excessive dimensional variation of the product using dimensional measurements. For such problems, the challenge is solving an underdetermined system caused by a common phenomenon in practice; namely, the number of measurements is less than that of the process errors. To address this challenge, this paper attempts to solve the following two problems: (1) how to utilize the temporal correlation in the time series data of each process error and (2) how to apply prior knowledge regarding which process errors are more likely to be process faults. A novel sparse Bayesian learning method is proposed to achieve the above objectives. The method consists of three hierarchical layers. The first layer has parameterized prior distribution that exploits the temporal correlation of each process error. Furthermore, the second and third layers achieve the prior distribution representing the prior knowledge of process faults. Then, these prior distributions are updated with the likelihood function of the measurement samples from the process, resulting in the accurate posterior distribution of process faults from an underdetermined system. Since posterior distributions of process faults are intractable, this paper derives approximate posterior distributions via Variational Bayes inference. Numerical and simulation case studies using an actual autobody assembly process are performed to demonstrate the effectiveness of the proposed method.

stat.AP↗

arXiv2022

Reinforcement Learning-based Defect Mitigation for Quality Assurance of Additive Manufacturing

Additive Manufacturing (AM) is a powerful technology that produces complex 3D geometries using various materials in a layer-by-layer fashion. However, quality assurance is the main challenge in AM industry due to the possible time-varying processing conditions during AM process. Notably, new defects may occur during printing, which cannot be mitigated by offline analysis tools that focus on existing defects. This challenge motivates this work to develop online learning-based methods to deal with the new defects during printing. Since AM typically fabricates a small number of customized products, this paper aims to create an online learning-based strategy to mitigate the new defects in AM process while minimizing the number of samples needed. The proposed method is based on model-free Reinforcement Learning (RL). It is called Continual G-learning since it transfers several sources of prior knowledge to reduce the needed training samples in the AM process. Offline knowledge is obtained from literature, while online knowledge is learned during printing. The proposed method develops a new algorithm for learning the optimal defect mitigation strategies proven the best performance when utilizing both knowledge sources. Numerical and real-world case studies in a fused filament fabrication (FFF) platform are performed and demonstrate the effectiveness of the proposed method.

cs.LG↗

arXiv2022

Rational Uniform Consensus with General Omission Failures

Generally, system failures, such as crash failures, Byzantine failures and so on, are considered as common reasons for the inconsistencies of distributed consensus and have been extensively studied. In fact, strategic manipulations by rational agents do not be ignored for reaching consensus in distributed system. In this paper, we extend the game-theoretic analysis of consensus and design an algorithm of rational uniform consensus with general omission failures under the assumption that processes are controlled by rational agents and prefer consensus. Different from crashing one, agent with omission failures may crash, or omit to send or receive messages when it should, which leads to difficulty of detecting faulty agents. By combining the possible failures of agents at the both ends of a link, we convert omission failure model into link state model to make faulty detection possible. Through analyzing message passing mechanism in the distributed system with n agents, among which t agents may commit omission failures, we provide the upper bound on message passing time for reaching consensus on a state among nonfaulty agents, and message chain mechanism for validating messages. And then we prove our rational uniform consensus is a Nash equilibrium when n>2t+1, and failure patterns and initial preferences are blind (an assumption of randomness). Thus agents could have no motivation to deviate the consensus. Our research strengthens the reliability of consensus with omission failures from the perspective of game theory.

cs.GT↗

arXiv2022

PanGu-Coder: Program Synthesis with Function-Level Language Modeling

We present PanGu-Coder, a pretrained decoder-only language model adopting the PanGu-Alpha architecture for text-to-code generation, i.e. the synthesis of programming language solutions given a natural language problem description. We train PanGu-Coder using a two-stage strategy: the first stage employs Causal Language Modelling (CLM) to pre-train on raw programming language data, while the second stage uses a combination of Causal Language Modelling and Masked Language Modelling (MLM) training objectives that focus on the downstream task of text-to-code generation and train on loosely curated pairs of natural language program definitions and code functions. Finally, we discuss PanGu-Coder-FT, which is fine-tuned on a combination of competitive programming problems and code with continuous integration tests. We evaluate PanGu-Coder with a focus on whether it generates functionally correct programs and demonstrate that it achieves equivalent or better performance than similarly sized models, such as CodeX, while attending a smaller context window and training on less data.

cs.LG↗

arXiv2022

Self-scalable Tanh (Stan): Faster Convergence and Better Generalization in Physics-informed Neural Networks

Physics-informed Neural Networks (PINNs) are gaining attention in the engineering and scientific literature for solving a range of differential equations with applications in weather modeling, healthcare, manufacturing, etc. Poor scalability is one of the barriers to utilizing PINNs for many real-world problems. To address this, a Self-scalable tanh (Stan) activation function is proposed for the PINNs. The proposed Stan function is smooth, non-saturating, and has a trainable parameter. During training, it can allow easy flow of gradients to compute the required derivatives and also enable systematic scaling of the input-output mapping. It is shown theoretically that the PINNs with the proposed Stan function have no spurious stationary points when using gradient descent algorithms. The proposed Stan is tested on a number of numerical studies involving general regression problems. It is subsequently used for solving multiple forward problems, which involve second-order derivatives and multiple dimensions, and an inverse problem where the thermal diffusivity of a rod is predicted with heat conduction data. These case studies establish empirically that the Stan activation function can achieve better training and more accurate predictions than the existing activation functions in the literature.

cs.LG↗

arXiv2022

Smooth Robust Tensor Completion for Background/Foreground Separation with Missing Pixels: Novel Algorithm with Convergence Guarantee

The objective of this study is to address the problem of background/foreground separation with missing pixels by combining the video acquisition, video recovery, background/foreground separation into a single framework. To achieve this, a smooth robust tensor completion (SRTC) model is proposed to recover the data and decompose it into the static background and smooth foreground, respectively. Specifically, the static background is modeled by the low-rank tucker decomposition and the smooth foreground (moving objects) is modeled by the spatiotemporal continuity, which is enforced by the total variation regularization. An efficient algorithm based on tensor proximal alternating minimization (tenPAM) is implemented to solve the proposed model with global convergence guarantee under very mild conditions. Extensive experiments on real data demonstrate that the proposed method significantly outperforms the state-of-the-art approaches for background/foreground separation with missing pixels.

cs.CV↗