Case-based Reasoning for Natural Language Queries over Knowledge Bases

By Rajarshi Das, Manzil Zaheer, Dung Ngoc Thai, Ameya Godbole, Ethan Perez, Jay Yoon Lee, Lizhen Tan, Lazaros C. Polymenakos, Andrew McCallum. 2021

212 citations recorded by the source corpus.

It is often challenging to solve a complex problem from scratch, but much easier if we can access other similar problems with their solutions — a paradigm known as case-based reasoning (CBR). We propose a neuro-symbolic CBR approach (CBR-KBQA) for question answering over large knowledge bases. CBR-KBQA consists of a nonparametric memory that stores cases (question and logical forms) and a parametric model that can generate a logical form for a new question by retrieving cases that are relevant to it. On several KBQA datasets that contain complex questions, CBR-KBQA achieves competitive performance. For example, on the CWQ dataset, CBR-KBQA outperforms the current state of the art by 11% on accuracy. Furthermore, we show that CBR-KBQA is capable of using new cases without any further training: by incorporating a few human-labeled examples in the case memory, CBR-KBQA is able to successfully generate logical forms containing unseen KB entities as well as relations.

Subfields: Artificial Intelligence, Natural Language Processing, Knowledge Graphs & Representation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Knowledge Graphs & Representation

Jonathan Berant

Citations in the corpus

Learning To Retrieve Prompts for In-Context Learning (2021, methodology): “• C ASE - BASED R EASONING (CBR): We adapt the scoring function from Das et al. (2021), which focused on knowledge-base question answering.”
Case-Based Abductive Natural Language Inference (2020, methodology): “In contrast with the dominant paradigm, we propose to integrate abductive natural language inference in a Case-Based Reasoning (CBR) framework (Schank et al., 2014; Das et al., 2021).”
Retrieval-Augmented Generation for AI-Generated Content: A Survey (2024, unknown): “CBR-KBQA [54] revises the result by aligning generated relations with those presented in the local neighborhood of the query entity in knowledge graph.”
RNG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering (2021, background): “Similarly, we initiate the logical form ranker us-F1 EM Hits @1 PullNet* (Sun et al., 2019) 62.8 − 67.8 GraftNet* (Sun et al., 2018) − − 68.1 Bert Ranking* (Gu et al., 2021) 67.0 − − EmbedQA* (Saxena et al., 2020) − − 72.5 ReTrack* (Chen et al., 2021) 74.7 − 74.6 Topic Units (Lan et al., 2019) 67.9 ”
Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph (2023, unknown): “One of the main advantages of ToG is its plug-and-play capabilities.”
LM-VC: Zero-Shot Voice Conversion via Speech Generation Based on Language Models (2023, background): “Meanwhile, leveraging longer prompt more effectively [40] is also a promising solution to break such length restriction.”