Natural Language Processing65 [2025-1] 정유림 - GNN (GCN, GraphSAGE, GAT) PaperGCN : https://arxiv.org/abs/1609.02907GraphSAGE :https://arxiv.org/abs/1706.02216GAT : https://arxiv.org/pdf/1710.10903 1. 그래프 데이터 기본 개념그래프는 노드(Vertex)와 엣지(Edge) 로 이루어지며, 이를 수학적으로 다음과 같이 표현합니다.인접 행렬 (Adjacency Matrix), 노드 간 연결 관계를 나타냄.노드의 특징 행렬 (Feature Matrix), 각 노드의 feature 값을 포함. (초기 Feature가 GNN을 거치면서 학습되고, 최종적으로 Embedding이 됩니다.) 2. 그래프로 표현할 수 있는 데이터 예시분자 구조: 원자들이 노드, 결합이 엣지로 표현됨.소셜 네트.. 2025. 3. 16. [2025-1] 백승우 - Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Insights from Multi-Agent Collaboration Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Insights from Multi-Agent CollaborationWeikang Yuan, Junjie Cao, Zhuoren Jiang, Yangyang Kang, Jun Lin, Kaisong Song, Tianqianjin Lin, Pengwei Yan, Changlong Sun, Xiaozhong Liu. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.aclanthology.orgMotivationsLegal 분야에서는 LLMs를 이용해서 법 이론을 충분히 이해하고 복잡.. 2025. 3. 7. [2025-1] 현시은 - PlanRAG: A Plan-then-Retrieval Augmented Generation for Generative Large Language Models as Decision Makers 원본 논문 링크 : https://arxiv.org/abs/2406.12430 PlanRAG: A Plan-then-Retrieval Augmented Generation for Generative Large Language Models as Decision MakersIn this paper, we conduct a study to utilize LLMs as a solution for decision making that requires complex data analysis. We define Decision QA as the task of answering the best decision, $d_{best}$, for a decision-making question $Q$, business rul.. 2025. 3. 6. [2025-1] 백승우 - A-MEM: Agentic Memory for LLM Agents A-MEM: Agentic Memory for LLM AgentsWhile large language model (LLM) agents can effectively use external tools for complex real-world tasks, they require memory systems to leverage historical experiences. Current memory systems enable basic storage and retrieval but lack sophisticated memoryarxiv.org1. IntroductionLLM agent의 발전으로, 환경과 상호작용하고 작업을 실행하며 의사결정을 할 수 있게됨Reasoning과 planning 능력을 향상시키기 위해.. 2025. 3. 5. 이전 1 2 3 4 5 ··· 17 다음
