Incremental Knowledge Based Question Answering

2021-01-18 · Yongqi Li, Wenjie Li, Liqiang Nie

In the past years, Knowledge-Based Question Answering (KBQA), which aims to answer natural language questions using facts in a knowledge base, has been well developed. Existing approaches often assume a static knowledge base. However, the knowledge is evolving over time in the real world. If we directly apply a fine-tuning strategy on an evolving knowledge base, it will suffer from a serious catastrophic forgetting problem. In this paper, we propose a new incremental KBQA learning framework that can progressively expand learning capacity as humans do. Specifically, it comprises a margin-distilled loss and a collaborative exemplar selection method, to overcome the catastrophic forgetting problem by taking advantage of knowledge distillation. We reorganize the SimpleQuestion dataset to evaluate the proposed incremental learning solution to KBQA. The comprehensive experiments demonstrate its effectiveness and efficiency when working with the evolving knowledge base.

📄 PDF Abstract BibTeX arXiv:2101.06938

Code (0)

등록된 구현이 없습니다.

Tasks

Incremental Learning Knowledge Distillation Question Answering

Incremental Reading for Question Answering

2019-01-15 · Samira Abnar, Tania Bedrax-Weiss, Tom Kwiatkowski, William W. Cohen

Any system which performs goal-directed continual learning must not only learn incrementally but process and absorb information incrementally. Such a system also has to understand when its goals have been achieved. In th…

Continual Learning Question Answering

Quizbowl: The Case for Incremental Question Answering

2019-04-09 · Pedro Rodriguez, Shi Feng, Mohit Iyyer, He He 외

Scholastic trivia competitions test knowledge and intelligence through mastery of question answering. Modern question answering benchmarks are one variant of the Turing test. Specifically, answering a set of questions as…

BIG-bench Machine Learning Decision Making Question Answering Sequential Decision Making

Task Formulation Matters When Learning Continuously: A Case Study in Visual Question Answering

2022-01-16 · ACL ARR January 2022 1 · Anonymous

Continual learning is a promising alternative to the current pretrain-and-finetune paradigm: It aims to learn a model on a sequence of tasks without forgetting knowledge from preceding tasks. We investigate continual lea…

Continual Learning Incremental Learning Question Answering Semantic Similarity+3

Mining Object Parts from CNNs via Active Question-Answering

2017-04-11 · CVPR 2017 7 · Quanshi Zhang, Ruiming Cao, Ying Nian Wu, Song-Chun Zhu

Given a convolutional neural network (CNN) that is pre-trained for object classification, this paper proposes to use active question-answering to semanticize neural patterns in conv-layers of the CNN and mine part concep…

Active Learning Object Question Answering

KBQA-o1: Agentic Knowledge Base Question Answering with Monte Carlo Tree Search

2025-01-31 · Haoran Luo, Haihong E, Yikai Guo, Qika Lin 외

Knowledge Base Question Answering (KBQA) aims to answer natural language questions with a large-scale structured knowledge base (KB). Despite advancements with large language models (LLMs), KBQA still faces challenges in…

Heuristic Search Knowledge Base Question Answering Question Answering