It is well acknowledged that incorporating explicit knowledge graphs (KGs) can benefit question answering. Existing approaches typically follow a grounding-reasoning pipeline in which entity nodes are first grounded for the query (question and candidate answers), and then a reasoning module reasons over the matched multi-hop subgraph for answer prediction. Although the pipeline largely alleviates the issue of extracting essential information from giant KGs, efficiency is still an open challenge when scaling up hops in grounding the subgraphs. In this paper, we target at finding semantically related entity nodes in the subgraph to improve the efficiency of graph reasoning with KG. We propose a grounding-pruning-reasoning pipeline to prune noisy nodes, remarkably reducing the computation cost and memory usage while also obtaining decent subgraph representation. In detail, the pruning module first scores concept nodes based on the dependency distance between matched spans and then prunes the nodes according to score ranks. To facilitate the evaluation of pruned subgraphs, we also propose a graph attention network (GAT) based module to reason with the subgraph data. Experimental results on CommonsenseQA and OpenBookQA demonstrate the effectiveness of our method.
Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search
University of Science and Technology of China（中国科学技术大学）；Institute of Artificial Intelligence, Hefei Comprehensive National Science Center（合肥综合国家科学中心人工智能研究院）
The Benchmark Illusion: Pruned LLMs Can Pass Multiple Choice but Fail to Answer
Institute of Science Tokyo（东京科学大学）；Tohoku University（东北大学）；Nanyang Technological University（南洋理工大学）；KTH Royal Institute of Technology（瑞典皇家理工学院）；Shandong University（山东大学）
Learning to Focus: Causal Attention Distillation via Gradient-Guided Token Pruning
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；Department of Computer Science and Technology, Tsinghua University（清华大学计算机科学与技术系）；Baidu Inc.（百度公司）
MoDora: Tree-Based Semi-Structured Document Analysis System
Shanghai Jiao Tong University（上海交通大学）；Microsoft Research（微软研究院）；Beihang University（北航）；Shanghai AI Lab（上海人工智能实验室）；Tsinghua University（清华大学）；Institute for Clarity in Documentation（文档清晰性研究所）；Inria Paris-Rocquencourt（Inria巴黎-罗quentcourt研究中心）；East China Normal University（华东师范大学）；Scientific Writing Academy（科学写作学院）
Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳））；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）；Tsinghua University（清华大学）；The Chinese University of Hong Kong（香港中文大学）；Zhongguancun Academy（中关村学院）；Shenzhen Loop Area Institute（深圳河套学院）
When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale Tables
University of Science and Technology of China（中国科学技术大学）；The University of Melbourne（墨尔本大学）；Institute of Artificial Intelligence, Hefei Comprehensive National Science Center（合肥综合性国家科学中心人工智能研究院）
Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents
University of Science and Technology of China（中国科学技术大学）；National University of Singapore（新加坡国立大学）；Shanghai Jiao Tong University（上海交通大学）；DP Technology（DP科技）；Meituan（美团）
Search-on-Graph: Iterative Informed Navigation for Large Language Model Reasoning on Knowledge Graphs
Université de Montréal（蒙特利尔大学）；McGill University（麦吉尔大学）；Digital Media, CBC（数字媒体，CBC）；Halmstad University College（哈姆斯塔德大学学院）；University of Waterloo（滑铁卢大学）
