Taking Up the Gaokao Challenge: An Information Retrieval Approach

Authors: Gong Cheng, Weixi Zhu, Ziwei Wang, Jianghui Chen, Yuzhong Qu

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our approach achieves encouraging results on real-life questions in recent history tests, significantly outperforming baseline approaches.
Researcher Affiliation Academia National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, China
Pseudocode Yes Algorithm 1: Retrieving Concept Pages
Open Source Code No The paper states: 'We have made our dataset online accessible to the research community3.' with a link to the dataset, but no explicit statement or link for the source code of the methodology itself.
Open Datasets Yes We have made our dataset online accessible to the research community3. 3http://ws.nju.edu.cn/gaokao/ijcai-16/GaokaoHistory577.xml
Dataset Splits No The paper describes its dataset split into QS-A (123 questions) and QS-B (454 questions) used for evaluation. However, it does not provide explicit training, validation, or test dataset splits in terms of percentages, sample counts, or predefined partition files.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, processor types, memory amounts) used for running the experiments are provided in the paper.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., programming languages, libraries, frameworks, or specific solvers).
Experiment Setup Yes The parameters α = 0.8, β = 0.5, γ = 1.0 are empirically set in our experiments (Equation 1). The approach was configured to consistently use k = 6 top-ranked pages for answering every question.