PyTorchKR Papers

“Code Generation”

총 10,140편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2406.12837공식 코드 · Papers with Code

LayerMerge: Neural Network Depth Compression through Layer Pruning and Merging

Recent works show that reducing the number of layers in a convolutional neural network can enhance efficiency while maintaining the performance of the network. Existing depth compression methods remove redundant non-linear activation functions and merge the consecutive convolution layers into a single layer. However, these methods suffer from a critical drawback; the kernel size of the merged layers becomes larger, significantly undermining the latency reduction gained from reducing the depth of the network. We show that this problem can be addressed by jointly pruning convolution layers and activation functions. To this end, we propose LayerMerge, a novel depth compression method that selects which activation layers and convolution layers to remove, to achieve a desired inference speed-up while minimizing performance loss. Since the corresponding selection problem involves an exponential search space, we formulate a novel surrogate optimization problem and efficiently solve it via dynamic programming. Empirical results demonstrate that our method consistently outperforms existing depth compression and layer pruning methods on various network architectures, both on image classification and generation tasks. We release the code at https://github.com/snu-mllab/LayerMerge.

arXiv:2406.19749공식 코드 · Papers with Code

SPIRONet: Spatial-Frequency Learning and Topological Channel Interaction Network for Vessel Segmentation

Automatic vessel segmentation is paramount for developing next-generation interventional navigation systems. However, current approaches suffer from suboptimal segmentation performances due to significant challenges in intraoperative images (i.e., low signal-to-noise ratio, small or slender vessels, and strong interference). In this paper, a novel spatial-frequency learning and topological channel interaction network (SPIRONet) is proposed to address the above issues. Specifically, dual encoders are utilized to comprehensively capture local spatial and global frequency vessel features. Then, a cross-attention fusion module is introduced to effectively fuse spatial and frequency features, thereby enhancing feature discriminability. Furthermore, a topological channel interaction module is designed to filter out task-irrelevant responses based on graph neural networks. Extensive experimental results on several challenging datasets (CADSA, CAXF, DCA1, and XCAD) demonstrate state-of-the-art performances of our method. Moreover, the inference speed of SPIRONet is 21 FPS with a 512x512 input size, surpassing clinical real-time requirements (6~12FPS). These promising outcomes indicate SPIRONet's potential for integration into vascular interventional navigation systems. Code is available at https://github.com/Dxhuang-CASIA/SPIRONet.

arXiv:2410.12694공식 코드 · Papers with Code

VividMed: Vision Language Model with Versatile Visual Grounding for Medicine

Recent advancements in Vision Language Models (VLMs) have demonstrated remarkable promise in generating visually grounded responses. However, their application in the medical domain is hindered by unique challenges. For instance, most VLMs rely on a single method of visual grounding, whereas complex medical tasks demand more versatile approaches. Additionally, while most VLMs process only 2D images, a large portion of medical images are 3D. The lack of medical data further compounds these obstacles. To address these challenges, we present VividMed, a vision language model with versatile visual grounding for medicine. Our model supports generating both semantic segmentation masks and instance-level bounding boxes, and accommodates various imaging modalities, including both 2D and 3D data. We design a three-stage training procedure and an automatic data synthesis pipeline based on open datasets and models. Besides visual grounding tasks, VividMed also excels in other common downstream tasks, including Visual Question Answering (VQA) and report generation. Ablation studies empirically show that the integration of visual grounding ability leads to improved performance on these tasks. Our code is publicly available at https://github.com/function2-llx/MMMM.

arXiv:2502.01860

SE Arena: An Interactive Platform for Evaluating Foundation Models in Software Engineering

Foundation models (FMs), particularly large language models (LLMs), have shown significant promise in various software engineering (SE) tasks, including code generation, debugging, and requirement refinement. Despite these advances, existing evaluation frameworks are insufficient for assessing model performance in iterative, context-rich workflows characteristic of SE activities. To address this limitation, we introduce SE Arena, an interactive platform designed to evaluate FMs in SE tasks. SE Arena provides a transparent, open-source leaderboard, supports multi-round conversational workflows, and enables end-to-end model comparisons. The platform introduces novel metrics, including model consistency score that measures the consistency of model outputs through self-play matches, and conversation efficiency index that evaluates model performance while accounting for the number of interaction rounds required to reach conclusions. Moreover, SE Arena incorporates a new feature called RepoChat, which automatically injects repository-related context (e.g., issues, commits, pull requests) into the conversation, further aligning evaluations with real-world development processes. This paper outlines the design and capabilities of SE Arena, emphasizing its potential to advance the evaluation and practical application of FMs in software engineering.

arXiv:2502.11901

Building A Proof-Oriented Programmer That Is 64% Better Than GPT-4o Under Data Scarcity

Existing LMs struggle with proof-oriented programming due to data scarcity, which manifest in two key ways: (1) a lack of sufficient corpora for proof-oriented programming languages such as F*, and (2) the absence of large-scale, project-level proof-oriented implementations that can teach the model the intricate reasoning process when performing proof-oriented programming. We present the first on synthetic data augmentation for project level proof oriented programming for both generation and repair. Our method addresses data scarcity by synthesizing basic proof-oriented programming problems for proficiency in that language; incorporating diverse coding data for reasoning capability elicitation and creating new proofs and repair data within existing repositories. This approach enables language models to both synthesize and repair proofs for function- and repository-level code. We show that our fine-tuned 14B parameter model, PoPilot, can exceed the performance of the models that outperforms GPT-4o in project-level proof-oriented programming by 64% relative margin, and can improve GPT-4o's performance by 54% by repairing its outputs over GPT-4o's self-repair.

arXiv:2504.08852

ML For Hardware Design Interpretability: Challenges and Opportunities

The increasing size and complexity of machine learning (ML) models have driven the growing need for custom hardware accelerators capable of efficiently supporting ML workloads. However, the design of such accelerators remains a time-consuming process, heavily relying on engineers to manually ensure design interpretability through clear documentation and effective communication. Recent advances in large language models (LLMs) offer a promising opportunity to automate these design interpretability tasks, particularly the generation of natural language descriptions for register-transfer level (RTL) code, what we refer to as "RTL-to-NL tasks." In this paper, we examine how design interpretability, particularly in RTL-to-NL tasks, influences the efficiency of the hardware design process. We review existing work adapting LLMs for these tasks, highlight key challenges that remain unaddressed, including those related to data, computation, and model development, and identify opportunities to address them. By doing so, we aim to guide future research in leveraging ML to automate RTL-to-NL tasks and improve hardware design interpretability, thereby accelerating the hardware design process and meeting the increasing demand for custom hardware accelerators in machine learning and beyond.

arXiv:2504.10041공식 코드 · Papers with Code

Prior Does Matter: Visual Navigation via Denoising Diffusion Bridge Models

Recent advancements in diffusion-based imitation learning, which show impressive performance in modeling multimodal distributions and training stability, have led to substantial progress in various robot learning tasks. In visual navigation, previous diffusion-based policies typically generate action sequences by initiating from denoising Gaussian noise. However, the target action distribution often diverges significantly from Gaussian noise, leading to redundant denoising steps and increased learning complexity. Additionally, the sparsity of effective action distributions makes it challenging for the policy to generate accurate actions without guidance. To address these issues, we propose a novel, unified visual navigation framework leveraging the denoising diffusion bridge models named NaviBridger. This approach enables action generation by initiating from any informative prior actions, enhancing guidance and efficiency in the denoising process. We explore how diffusion bridges can enhance imitation learning in visual navigation tasks and further examine three source policies for generating prior actions. Extensive experiments in both simulated and real-world indoor and outdoor scenarios demonstrate that NaviBridger accelerates policy inference and outperforms the baselines in generating target action sequences. Code is available at https://github.com/hren20/NaiviBridger.

arXiv:2504.11669

Co-STAR: Collaborative Curriculum Self-Training with Adaptive Regularization for Source-Free Video Domain Adaptation

Recent advances in Source-Free Unsupervised Video Domain Adaptation (SFUVDA) leverage vision-language models to enhance pseudo-label generation. However, challenges such as noisy pseudo-labels and over-confident predictions limit their effectiveness in adapting well across domains. We propose Co-STAR, a novel framework that integrates curriculum learning with collaborative self-training between a source-trained teacher and a contrastive vision-language model (CLIP). Our curriculum learning approach employs a reliability-based weight function that measures bidirectional prediction alignment between the teacher and CLIP, balancing between confident and uncertain predictions. This function preserves uncertainty for difficult samples, while prioritizing reliable pseudo-labels when the predictions from both models closely align. To further improve adaptation, we propose Adaptive Curriculum Regularization, which modifies the learning priority of samples in a probabilistic, adaptive manner based on their confidence scores and prediction stability, mitigating overfitting to noisy and over-confident samples. Extensive experiments across multiple video domain adaptation benchmarks demonstrate that Co-STAR consistently outperforms state-of-the-art SFUVDA methods. Code is available at: https://github.com/Plrbear/Co-Star

arXiv:2505.11853공식 코드 · Papers with Code

Measurement Score-Based Diffusion Model

Diffusion models are widely used in applications ranging from image generation to inverse problems. However, training diffusion models typically requires clean ground-truth images, which are unavailable in many applications. We introduce the Measurement Score-based diffusion Model (MSM), a novel framework that learns partial measurement scores using only noisy and subsampled measurements. MSM models the distribution of full measurements as an expectation over partial scores induced by randomized subsampling. To make the MSM representation computationally efficient, we also develop a stochastic sampling algorithm that generates full images by using a randomly selected subset of partial scores at each step. We additionally propose a new posterior sampling method for solving inverse problems that reconstructs images using these partial scores. We provide a theoretical analysis that bounds the Kullback-Leibler divergence between the distributions induced by full and stochastic sampling, establishing the accuracy of the proposed algorithm. We demonstrate the effectiveness of MSM on natural images and multi-coil MRI, showing that it can generate high-quality images and solve inverse problems -- all without access to clean training data. Code is available at https://github.com/wustl-cig/MSM.

arXiv:2505.18777

HD-PiSSA: High-Rank Distributed Orthogonal Adaptation

Existing parameter-efficient fine-tuning (PEFT) methods for large language models (LLMs), such as LoRA and PiSSA, constrain model updates to low-rank subspaces, limiting their expressiveness and leading to suboptimal performance on complex tasks. To address this, we introduce High-rank Distributed PiSSA (HD-PiSSA), a distributed PEFT approach that initializes orthogonal adapters across different devices and aggregates their delta updates collectively on W for fine-tuning. Unlike Data Parallel LoRA or PiSSA, which maintain identical adapters across all devices, HD-PiSSA assigns different principal components of the pre-trained weights to each GPU, significantly expanding the range of update directions. This results in over 16x higher effective updated ranks than data-parallel LoRA or PiSSA when fine-tuning on 8 GPUs with the same per-device adapter rank. Empirically, we evaluate HD-PiSSA across various challenging downstream tasks, including mathematics, code generation, and multi-task learning. In the multi-task setting, HD-PiSSA achieves average gains of 10.0 absolute points (14.63%) over LoRA and 4.98 points (6.60%) over PiSSA across 12 benchmarks, demonstrating its benefits from the extra optimization flexibility.

arXiv:2506.01822공식 코드 · Papers with Code

GSCodec Studio: A Modular Framework for Gaussian Splat Compression

3D Gaussian Splatting and its extension to 4D dynamic scenes enable photorealistic, real-time rendering from real-world captures, positioning Gaussian Splats (GS) as a promising format for next-generation immersive media. However, their high storage requirements pose significant challenges for practical use in sharing, transmission, and storage. Despite various studies exploring GS compression from different perspectives, these efforts remain scattered across separate repositories, complicating benchmarking and the integration of best practices. To address this gap, we present GSCodec Studio, a unified and modular framework for GS reconstruction, compression, and rendering. The framework incorporates a diverse set of 3D/4D GS reconstruction methods and GS compression techniques as modular components, facilitating flexible combinations and comprehensive comparisons. By integrating best practices from community research and our own explorations, GSCodec Studio supports the development of compact representation and compression solutions for static and dynamic Gaussian Splats, namely our Static and Dynamic GSCodec, achieving competitive rate-distortion performance in static and dynamic GS compression. The code for our framework is publicly available at https://github.com/JasonLSC/GSCodec_Studio , to advance the research on Gaussian Splats compression.

arXiv:2506.00385공식 코드 · Papers with Code

MagiCodec: Simple Masked Gaussian-Injected Codec for High-Fidelity Reconstruction and Generation

Neural audio codecs have made significant strides in efficiently mapping raw audio waveforms into discrete token representations, which are foundational for contemporary audio generative models. However, most existing codecs are optimized primarily for reconstruction quality, often at the expense of the downstream modelability of the encoded tokens. Motivated by the need to overcome this bottleneck, we introduce $\textbf{MagiCodec}$, a novel single-layer, streaming Transformer-based audio codec. MagiCodec is designed with a multistage training pipeline that incorporates Gaussian noise injection and latent regularization, explicitly targeting the enhancement of semantic expressiveness in the generated codes while preserving high reconstruction fidelity. We analytically derive the effect of noise injection in the frequency domain, demonstrating its efficacy in attenuating high-frequency components and fostering robust tokenization. Extensive experimental evaluations show that MagiCodec surpasses state-of-the-art codecs in both reconstruction quality and downstream tasks. Notably, the tokens produced by MagiCodec exhibit Zipf-like distributions, as observed in natural languages, thereby improving compatibility with language-model-based generative architectures. The code and pre-trained models are available at https://github.com/Ereboas/MagiCodec.

arXiv:2506.03230공식 코드 · Papers with Code

DiaBlo: Diagonal Blocks Are Sufficient For Finetuning

Finetuning is a critical step for adapting large language models (LLMs) to domain-specific downstream tasks. To mitigate the substantial computational and memory costs of full-model fine-tuning, Parameter-Efficient Finetuning (PEFT) methods have been proposed to update only a small subset of model parameters. However, performance gaps between PEFT approaches and full-model fine-tuning still exist. In this work, we present DiaBlo, a simple yet effective PEFT approach that updates only the diagonal blocks of selected model weight matrices. Unlike Low Rank Adaptation (LoRA) and its variants, DiaBlo eliminates the need for low rank matrix products, thereby avoiding the reliance on auxiliary initialization schemes or customized optimization strategies to improve convergence. This design leads to stable and robust convergence while maintaining comparable memory efficiency and training speed to LoRA. We conduct extensive experiments across a range of tasks, including commonsense reasoning, arithmetic reasoning, code generation, and safety alignment, to evaluate the effectiveness and efficiency of DiaBlo. Across these benchmarks, DiaBlo demonstrates strong and consistent performance while maintaining high memory efficiency and fast finetuning speed. Codes are available at https://github.com/ziyangjoy/DiaBlo.

arXiv:2506.15263공식 코드 · Papers with Code

Minimizing Structural Vibrations via Guided Flow Matching Design Optimization

Structural vibrations are a source of unwanted noise in engineering systems like cars, trains or airplanes. Minimizing these vibrations is crucial for improving passenger comfort. This work presents a novel design optimization approach based on guided flow matching for reducing vibrations by placing beadings (indentations) in plate-like structures. Our method integrates a generative flow matching model and a surrogate model trained to predict structural vibrations. During the generation process, the flow matching model pushes towards manufacturability while the surrogate model pushes to low-vibration solutions. The flow matching model and its training data implicitly define the design space, enabling a broader exploration of potential solutions as no optimization of manually-defined design parameters is required. We apply our method to a range of differentiable optimization objectives, including direct optimization of specific eigenfrequencies through careful construction of the objective function. Results demonstrate that our method generates diverse and manufacturable plate designs with reduced structural vibrations compared to designs from random search, a criterion-based design heuristic and genetic optimization. The code and data are available from https://github.com/ecker-lab/Optimizing_Vibrating_Plates.

arXiv:2510.06596v2cs.CV

SDQM: 객체 검출 데이터셋 평가를 위한 합성 데이터 품질 지표

SDQM: Synthetic Data Quality Metric for Object Detection Dataset Evaluation

머신러닝 모델의 성능은 학습 데이터에 크게 의존합니다. 대규모의 잘 주석이 달린 데이터셋 부족은 견고한 모델 생성에 있어 중요한 도전 과제를 야기합니다. 이를 해결하기 위해 시뮬레이션 및 생성 모델을 통해 생성된 합성 데이터가 유망한 해결책으로 등장하여, 데이터셋의 다양성을 높이고 모델의 성능, 신뢰성, 회복력을 향상시켰습니다. 그러나 이 생성된 데이터의 품질을 평가하려면 효과적인 지표가 필요합니다. 본 연구에서는 모델 학습(training)의 수렴이 필요 없이 객체 검출(object detection) 작업의 데이터 품질을 평가하기 위한 합성 데이터셋 품질 지표(SDQM)를 제안합니다. 이 지표는 합성 데이터셋의 생성 및 선택을 더욱 효율적으로 만들어, 자원이 제한된 객체 검출 작업의 핵심 과제를 해결합니다. 실험 결과, SDQM은 선도적인 객체 검출 모델인 YOLO11의 평균 정밀도(mAP) 점수와 강한 상관관계를 나타낸 반면, 기존 지표들은 중간 정도 또는 약한 상관관계만을 보였습니다. 또한, 비용이 많이 드는 반복적인 학습(iterative training)의 필요성을 최소화하면서 데이터셋 품질 개선을 위한 실행 가능한 통찰력을 제공합니다. 이 확장 가능하고 효율적인 지표는 합성 데이터 평가의 새로운 표준을 제시합니다. SDQM의 코드는 https://github.com/ayushzenith/SDQM 에서 확인할 수 있습니다.

arXiv:2606.18946v1cs.CL

SenFlow: 하이브리드 문서에서의 AI 생성 텍스트 탐지를 위한 문장 간 흐름 모델링

SenFlow: Inter-Sentence Flow Modeling for AI-Generated Text Detection in Hybrid Documents

하이브리드 문서의 문장 수준 AI 생성 텍스트 탐지(S-AGTD)를 문장 그래프 위 구조적 예측으로 재정의한 SenFlow는 그래프 기반 문장 간 전파와 선형 체인 CRF 디코딩을 단일 문서 패스로 결합한다. DeepSeek-V3.2·Kimi K2로 구축한 MOSAIC 벤치마크(16,000문서)에서 최고 성능을 보이며, 특히 교차 도메인 전이에서 Macro-F1이 평균 4.15pp 향상되고 혼란도 필터 이후에도 생성 모델에 따른 문장 길이 격차가 탐지 단서로 남는다.

arXiv:2509.25682v2cs.CV

퓨샷 합성 이미지 출처 추적: 제한된 샘플로 미확인 생성기 식별

Few-Shot Synthetic Image Attribution: Identifying Unseen Generators with Limited Samples

AI 생성 이미지(AIGI) 출처 추적은 단순한 AIGI 탐지를 넘어서는 시급한 과제로, 합성 이미지를 생성한 소스 모델이나 기법을 식별하는 것을 목표로 한다. 그러나 기존의 대부분 소스 출처 추적 방법은 폐쇄형(closed-set) 방식으로 동작하여 새로운 카테고리를 인식하기 위해서는 재학습이 필요하며, 이로 인해 빠르게 진화하는 이미지 생성 기술에 적응하기 어렵다. 본 연구에서는 퓨샷 출처 추적(few-shot attribution)이라는 새로운 패러다임을 제안한다. 이 패러다임은 제한된 샘플만을 사용하여 미확인 생성기를 신뢰성 있게 식별하는 것을 목표로 하며, 실제 응용 환경에 매우 적합하다. 이 연구를 지원하기 위해, 45개의 서로 다른 생성기로부터 수집한 117만 장의 이미지를 포함하는 대규모의 체계적으로 분류된 합성 이미지 데이터셋인 OmniFake를 구축하였다. 또한 이미지의 진위를 평가할 뿐만 아니라 합성 출처를 판별하는 퓨샷 출처 추적 기준선인 OmniDFA(Omni Detector and Few-shot Attributor)를 제안한다. 실험 결과, OmniDFA는 퓨샷 출처 추적에서 우수한 성능을 보이며 AIGI 탐지에서 최첨단 일반화 성능을 달성함을 입증하였다. 데이터셋과 코드는 https://github.com/teheperinko541/OmniDFA 에서 공개되어 있다.

arXiv:2607.08705v1cs.CV

HumanForge: 멀티 에이전트 위조 근거를 갖춘 인간 중심 딥페이크 영상 벤치마크

HumanForge: A Human-Centric Deepfake Video Benchmark with Multi-Agent Forgery Rationales

인간 중심 딥페이크(deepfake) 영상 탐지 연구의 한계를 극복하기 위해, 오디오 기반·포즈 기반·상호작용·의미 기반 등 네 가지 시나리오를 포괄하는 대규모 영상 위조 벤치마크 HumanForge(18,000개 이상의 합성 영상 세그먼트)를 제안한다. 데이터셋 구축 및 주석 자동화를 위해 LangGraph 기반의 멀티 에이전트 파이프라인 Gen2Anno를 설계하여, 생성 출처와 시각적 분석을 대조하는 방식으로 이진 판정·세밀한 아티팩트 범주·시공간적 위치 정보를 포함한 omni-annotation을 생성하고, 최신 탐지기 및 대규모 멀티모달 모델(LMM) 벤치마크를 통해 제로샷 일반화의 높은 난이도를 입증하였다.

arXiv:2607.14249v1cs.LG

MIDiff: 다변량-이미징 디퓨전을 통한 모바일 사용 생성의 희소성과 불균형 해결

MIDiff: Tackling Sparsity and Imbalance in Mobile Usage Generation via Multivariate-Imaging Diffusion

모바일 사용 추적 데이터는 사용자 행동 예측 및 앱 추천과 같은 작업에 중요하지만, 프라이버시 제약과 비용이 큰 대규모 데이터 수집으로 인해 활용이 제한된다. 생성 모델은 일반적인 시계열에서는 잘 작동하지만, 모바일 사용 데이터에 적용하기는 여전히 어렵다. 그 이유는 (i) 제한된 사용자 활동으로 인한 심각한 희소성, (ii) 이질적인 변수 유형으로 인한 공동 모델링의 복잡성, (iii) 앱 간 기능적 차이로 인한 뚜렷한 사용 불균형 때문이다. 이러한 과제를 해결하기 위해 Cross-Gramian Angular Sum Field(C-GASF)로 정의된 이미징 공간에서 동작하는 디퓨전 기반 프레임워크인 Multivariate-Imaging Diffusion(MIDiff)을 제안한다. C-GASF는 희소한 다변량 시퀀스를 상관 이미지로 변환하고, MIDiff는 U-Net에서 Triple 어텐션(attention)을 사용하여 시간적 일관성과 변수 간 의존성을 보존한다. 실험 결과 MIDiff는 충실도 지표 전반에서 최첨단 성능을 달성한다. 특히 Discriminative Accuracy(DA)는 0.1526을 기록했으며, 가장 강력한 기준선(baseline)인 ZITS-VAE의 0.3476과 비교된다. 이는 현실적이고 다양한 모바일 사용 추적 데이터 생성에 MIDiff가 효과적임을 보여준다. 코드는 https://github.com/YilaiLiu-HKU/MIDiff 에서 제공된다.

arXiv:2608.03600v2cs.AI

편미분방정식 워크플로를 위한 대규모 언어 모델

Large language models for partial differential equation workflows

수치 방법과 심층 학습은 편미분방정식(PDE)으로 기술되는 시스템의 모델링과 시뮬레이션을 발전시켜 왔지만, 방정식 수립, 솔버 구성, 시뮬레이션을 설계 및 제어 결정으로 전환하는 일은 여전히 상당한 도메인 전문성과 계산 자원을 요구한다. 대규모 언어 모델(LLM)은 과학적 지식, 수학적 추론, 코드 생성, 도구 사용을 결합하여 이러한 작업을 자동화하고 조율할 기회를 제공한다. 본 논문에서는 발견, 해결, 최적화를 포함한 세 단계에 걸친 LLM 보조 PDE 연구를 검토한다. LLM이 물리적 지식, 수치 방법, 계산 피드백과 어떻게 상호작용하여 모델을 수립하고, 해법 절차를 구현하며, 설계와 제어를 안내하는지 살펴본다. 또한 과학적 타당성, 계산 성능, 워크플로 신뢰성을 평가하기 위한 기존 벤치마크와 평가 관행을 검토한다. 마지막으로, 특정 PDE 문제에 맞춘 수학·물리 추론 개발, 도메인 특화 및 멀티모달 사후 학습을 통한 전이 가능한 도메인 전문성 구축, 개방형 문제를 위한 능동적 실험 설계와 과학적 검증의 통합이라는 상호 연결된 세 가지 과제와 기회를 논의한다. 이러한 방향은 PDE 연구 전반에서 과학적 아이디어와 계산 방법을 개발하고 검증하는 데 LLM이 보다 실질적인 역할을 수행할 수 있는 경로를 제시한다.