EMNLP Findings2021
Grounded Graph Decoding improves Compositional Generalization in Question Answering
Yu Gai, Paras Jain, Wendi Zhang, Joseph Gonzalez, Dawn Song, Ion Stoica
Abstract
Question answering models struggle to generalize to novel compositions of training patterns, such to longer sequences or more complex test structures. Current end-to-end models learn a flat input embedding which can lose input syntax context. Prior approaches improve generalization by learning permutation invariant models, but these methods do not scale to more complex train-test splits. We propose Grounded Graph Decoding, a method to improve compositional generalization of language representations by grounding structured predictions with an attention mechanism. Grounding enables the model to retain syntax information from the input in thereby significantly improving generalization over complex inputs. By predicting a structured graph containing conjunctions of query clauses, we learn a group invariant representation without making assumptions on the target domain. Our model significantly outperforms state-ofthe-art baselines on the Compositional Freebase Questions (CFQ) dataset, a challenging benchmark for compositional generalization in question answering. Moreover, we effectively solve the MCD1 split with 98% accuracy. All source is available at https:// github.com/gaiyu0/cfq.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
3 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Compositional Semantic Parsing with Large Language Models cites Learning Compositional Rules via Neural Program Synthesis
- Compositional Semantic Parsing with Large Language Models cites Measuring Compositional Generalization: A Comprehensive Method on Realistic Data
- Compositional Semantic Parsing with Large Language Models cites Compositional Generalization via Neural-Symbolic Stack Machines
- Compositional Semantic Parsing with Large Language Models cites Permutation Equivariant Models for Compositional Generalization in Language
- Compositional Generalization via Neural-Symbolic Stack Machines cites Learning Compositional Rules via Neural Program Synthesis
- Compositional Generalization via Neural-Symbolic Stack Machines cites Measuring Compositional Generalization: A Comprehensive Method on Realistic Data
- Compositional Generalization via Neural-Symbolic Stack Machines cites Permutation Equivariant Models for Compositional Generalization in Language
Cited by top-tier papers3
- Compositional Semantic Parsing with Large Language ModelsAndrew Drozdov, Nathanael Schärli, Ekin Akyürek, Nathan Scales et al.ICLR 2023 · 39 citations
- Bridging the gap to real-world language-grounded visual concept learningWhie Jung, Semin Kim, Junee Kim, Seunghoon HongNeurIPS 2025
- Latent Constraints on Unsupervised Text-Graph Alignment with Information AsymmetryJidong Tian, Wenqing Chen, Yitian Li, Caoyun Fan et al.AAAI 2023
Builds on4
- Measuring Compositional Generalization: A Comprehensive Method on Realistic DataDaniel Keysers, Nathanael Schärli, Nathan Scales, Hylke Buisman et al.ICLR 2020 · 401 citations
- Learning Compositional Rules via Neural Program SynthesisMaxwell I. Nye, Armando Solar-Lezama, Josh Tenenbaum, Brenden M. LakeNeurIPS 2020 · 120 citations
- Compositional Generalization via Neural-Symbolic Stack MachinesXinyun Chen, Chen Liang, Adams Wei Yu, Dawn Song et al.NeurIPS 2020 · 112 citations
- Permutation Equivariant Models for Compositional Generalization in LanguageJonathan Gordon, David Lopez-Paz, Marco Baroni, Diane BouchacourtICLR 2020 · 112 citations
Related papers
- Hierarchical Poset Decoding for Compositional Generalization in LanguageYinuo Guo, Zeqi Lin, Jian-Guang Lou, Dongmei ZhangNeurIPS 2020 · 34 citations
- Compositional Generalization with Grounded Language ModelsSondre Wold, Étienne Simon, Lucas Georges Gabriel Charpentier, Egor V. Kostylev et al.ACL Findings 2024
- Question Decomposition Tree for Answering Complex Questions over Knowledge BasesXiang Huang, Sitao Cheng, Yiheng Shu, Yuheng Bao et al.AAAI 2023 · 33 citations
- Compositional Generalization in Dependency ParsingEmily Goodwin, Siva Reddy, Timothy J. O'Donnell, Dzmitry BahdanauACL 2022
- Consistency of Compositional Generalization Across Multiple LevelsChuanhao Li, Zhen Li, Chenchen Jing, Xiaomeng Fan et al.AAAI 2025 · 1 citation
