Less is More: One-shot Subgraph Reasoning on Large-scale Knowledge Graphs
Hong Kong Baptist University(香港浸会大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));Shanghai Jiao Tong University(上海交通大学);Tsinghua University(清华大学)
Zhanke Zhou, Yongqi Zhang, Jiangchao Yao, Quanming Yao, Bo Han
To deduce new facts on a knowledge graph (KG), a link predictor learns from the graph structure and collects local evidence to find the answer to a given query. However, existing methods suffer from a severe scalability problem due to the utilization of the whole KG for prediction, which hinders their promise on large scale KGs and cannot be directly addressed by vanilla sampling methods. In this work, we propose the one-shot-subgraph link prediction to achieve efficient and adaptive prediction. The design principle is that, instead of directly acting on the whole KG, the prediction procedure is decoupled into two steps, i.e., (i) extracting only one subgraph according to the query and (ii) predicting on this single, query dependent subgraph. We reveal that the non-parametric and computation-efficient heuristics Personalized PageRank (PPR) can effectively identify the potential answers and supporting evidence. With efficient subgraph-based prediction, we further introduce the automated searching of the optimal configurations in both data and model spaces. Empirically, we achieve promoted efficiency and leading performances on five large-scale benchmarks. The code is publicly available at: https://github.com/tmlr-group/one-shot-subgraph.
Adapting Meta Knowledge Graph Information for Multi-Hop Reasoning over Few-Shot Relations
Tsinghua University(清华大学);Department of Computer Science and Technology, Tsinghua University(清华大学计算机科学与技术系);KIRC, Institute for Artificial Intelligence, Tsinghua University(清华大学人工智能研究院知识图谱与推理中心(KIRC));Beijing National Research Center for Information Science and Technology(北京国家信息科学与技术研究中心)
Context-aware Inductive Knowledge Graph Completion with Latent Type Constraints and Subgraph Reasoning
The Chinese University of Hong Kong(香港中文大学);International Digital Economy Academy(国际数字经济研究院);Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));University of Cambridge(剑桥大学)
xERTE: Explainable Reasoning on Temporal Knowledge Graphs for Forecasting Future Links
Institute of Informatics, LMU Munich(慕尼黑大学信息学研究所);Corporate Technology, Siemens AG(西门子股份公司中央研究院);Department of Informatics, Technical University of Munich(慕尼黑工业大学信息学系)
A Foundation Model for Zero-shot Logical Query Reasoning
Intel AI Lab(英特尔人工智能实验室);Purdue University(普渡大学);Mila - Québec AI Institute(米拉-魁北克人工智能研究所);HEC Montréal(蒙特利尔高等商学院);CIFAR AI Chair(加拿大高级研究院人工智能讲席);Université de Montréal(蒙特利尔大学)
Message Intercommunication for Inductive Relation Reasoning
School of Computer, National University of Defense Technology(国防科技大学计算机学院);College of Intelligence Science and Technology, National University of Defense Technology(国防科技大学智能科学学院)
ASP-based Multi-shot Reasoning via DLV2 with Incremental Grounding
University of Calabria(卡布里亚大学);Gruppo Nazionale Calcolo Scientifico-Istituto Nazionale di Alta Matematica(国家科学计算集团-国家高级数学研究所);DLVSystem Srl(DLVSystem公司)
Knowledge Base Question Answering by Case-based Reasoning over Subgraphs
UMass Amherst(马萨诸塞大学阿默斯特分校);University of Southern California(南加州大学);Google DeepMind(谷歌DeepMind);University of Washington(华盛顿大学)
知识库上的问答(QA)是一个具有挑战性的问题,因为它需要各种各样的推理模式。然而,我们假设在一个大的KB,推理模式需要回答的查询类型重新出现在各自的子图邻域中的各种实体。利用不同子图的局部邻域之间的这种结构相似性,我们引入了半参数模型(CBR-SUBG)具有(i)非参数分量,对于每个查询,动态检索其他类似的$k$ -最近邻(KNN)训练查询以及查询特定的子图,以及(ii)被训练以识别(潜在)从KNN查询的子图中推理模式,然后将它们应用于目标查询的子图。我们还提出了一个自适应的子图收集策略来选择一个查询特定的紧凑子图,使我们能够扩展到包含数十亿事实的完整Freebase KB。我们表明,CBR-SUBG可以回答需要子图推理模式的查询,并在几个KBQA基准测试中与最佳模型竞争。我们的子图收集策略还产生更紧凑的子图(例如,WebQSP的大小减少了55\%,而答案召回率增加了4.85\%)\footnote{Code, model, and subgraphs are available at \url{https://github.com/rajarshd/CBR-SUBG}}。
Beyond One-Size-Fits-All: Adaptive Subgraph Denoising for Zero-Shot Graph Learning with Large Language Models
JIUTIAN Research(中移九天);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));MIIT Key Laboratory of Data and Decision Intelligence(信息与决策智能重点实验室);Beihang University(北航)
Subgraph Retrieval Enhanced by Graph-Text Alignment for Commonsense Question Answering
School of Intelligence Science and Technology, Peking University(北京大学智能科学与技术学院);Ant Group(蚂蚁集团);School of Artificial Intelligence, Beijing Normal University(北京师范大学人工智能学院)
College of Computer Science and Technology, Zhejiang University of Technology(浙江工业大学计算机科学与技术学院);College of Computer Science and Technology, Zhejiang University(浙江大学计算机科学与技术学院)
Zero-Shot Visual Reasoning by Vision-Language Models: Benchmarking and Analysis
ASUS Intelligent Cloud Services (AICS)(华硕智能云服务(AICS));ASUS Global Pte. Ltd.(华硕全球私人有限公司);Centre for Frontier AI Research (CFAR)(前沿人工智能研究中心(CFAR));Institute of High Performance Computing (IHPC)(高性能计算研究所(IHPC));Agency for Science, Technology and Research (A*STAR)(新加坡科学、技术与研究局(A*STAR))
PathCoT: Chain-of-Thought Prompting for Zero-shot Pathology Visual Reasoning
The College of Artificial Intelligence, Nanjing University of Aeronautics(人工智能学院,南京航空航天大学);Astronautics The Key Laboratory of Brain-Machine Intelligence Technology, Ministry of Education(航空航天脑机智能技术重点实验室,教育部)
LLMs-based Few-Shot Disease Predictions using EHR: A Novel Approach Combining Predictive Agent Reasoning and Critical Agent Instruction
Emory University(埃默里大学);University of Washington(华盛顿大学);UCSD(加利福尼亚大学圣迭戈分校);Rollins School of Public Health, Emory University(埃默里大学罗林斯公共卫生学院);School of Medicine, Emory University(埃默里大学医学院)
K-level Reasoning for Zero-Shot Coordination in Hanabi
Facebook AI Research(脸书人工智能研究院);University of Oxford(牛津大学)
在合作的多代理设置的标准问题设置是自我发挥(SP),其目标是训练一个团队的代理一起工作。然而,最佳SP策略通常包含任意约定(“握手”),并且与其他独立训练的代理或人类不兼容。后一种需求最近由Hu等人(2020)正式定义为zero-shot协调(ZSC)设置,并部分通过他们的模拟游戏(OP)算法解决,该算法在纸牌游戏Hanabi中显示了改进的ZSC和人类AI性能。OP假设可以访问环境的对称性,并防止代理在训练期间以相互不兼容的方式破坏这些对称性。然而,正如作者所指出的,发现给定环境的对称性是一个计算上的难题。相反,我们表明,通过简单的k级推理(KLR)Costa Gomes et al. 2006,同步训练所有级别,我们可以在Hanabi中获得有竞争力的ZSC和ad-hoc团队合作性能,包括与类人代理机器人配对时。我们还介绍了一种新的方法,同步k级推理与最佳响应(SyKLRBR),它进一步提高了我们的同步KLR的性能,通过共同训练的最佳响应。