Hierarchical Poset Decoding for Compositional Generalization in Language

Authors: Yinuo Guo, Zeqi Lin, Jian-Guang Lou, Dongmei Zhang

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our proposed decoder on Compositional Freebase Questions (CFQ), a large and realistic natural language question answering dataset that is specifically designed to measure compositional generalization. Results show that it outperforms current decoders. (Abstract)
Researcher Affiliation Collaboration 1Key Laboratory of Computational Linguistics School of EECS, Peking University, Beijing, China; 2Microsoft Research Asia, Beijing, China
Pseudocode Yes Algorithm 1 generate_poset (the decoding process at inference time)
Open Source Code No The paper references external tools with links (GIZA++2, Match-Zoo3) but does not provide a link or explicit statement for the open-source code of their proposed method.
Open Datasets Yes Dataset We conduct experiments on three different splits MCD1/MCD2/MCD3 of CFQ dataset (Keysers et al., 2020).
Dataset Splits Yes Each split contains 95k/12k/12k instances for training/development/test. (Section 6.1)
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory, or cloud computing specifications) used for running the experiments.
Software Dependencies No The paper mentions software such as GIZA++2, Match-Zoo3, and PyTorch, but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes The dimensions of the hidden state and word embedding are set to 512, 300 respectively. The training process lasts 50 epochs with batch size 64/256 for sketch prediction and traversal path prediction, respectively. We use the Adam optimizer with default settings (in PYTORCH) and a dropout layer with the rate of 0.5. (Section 6.1)