ACL Findings2024
FRVA: Fact-Retrieval and Verification Augmented Entailment Tree Generation for Explainable Question Answering
Yue Fan, Hu Zhang, Ru Li, Yujie Wang, Hongye Tan, Jiye Liang
Abstract
Structured entailment tree can exhibit the reasoning chains from knowledge facts to predicted answers, which is important for constructing an explainable question answering system. Existing works mainly include directly generating the entire tree and stepwise generating the proof steps. The stepwise methods can exploit combinatoriality and generalize to longer steps, but they have large fact search spaces and error accumulation problems resulting in the generation of invalid steps. In this paper, inspired by the Dual Process Theory in cognitive science, we propose FRVA, a Fact-Retrieval and Verification Augmented bidirectional entailment tree generation method that contains two systems. Specifically, System 1 makes intuitive judgments through the fact retrieval module and filters irrelevant facts to reduce the search space. System 2 designs a deductive-abductive bidirectional reasoning module, and we construct cross-verification and multi-view contrastive learning to make the generated proof steps closer to the target hypothesis. We enhance the reliability of the stepwise proofs to mitigate error propagation. Experiment results on EntailmentBank show that FRVA outperforms previous models and achieves state-of-the-art performance in fact selection and structural correctness.
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.
2 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
- Leibniz: Theory-of-Mind Driven Neuro-Symbolic Logical Reasoning via Multi-Agent Collaboration cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Leibniz: Theory-of-Mind Driven Neuro-Symbolic Logical Reasoning via Multi-Agent Collaboration cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Self-Consistency Improves Chain of Thought Reasoning in Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Chain-of-Thought Prompting Elicits Reasoning in Large Language Models cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Deductive Verification of Chain-of-Thought Reasoning cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Deductive Verification of Chain-of-Thought Reasoning cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Cited by top-tier papers2
- Leibniz: Theory-of-Mind Driven Neuro-Symbolic Logical Reasoning via Multi-Agent CollaborationYue Fan, Hu Zhang, Yunxiao Zhao, Guangjun Zhang et al.ACL 2026
- Dynamic Energy-Based Contrastive Learning with Multi-Stage Knowledge Verification for Event Causality IdentificationYa Su, Hu Zhang, Yue Fan, Guangjun Zhang et al.EMNLP 2025
Builds on24
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- ALBERT: A Lite BERT for Self-supervised Learning of Language RepresentationsZhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel et al.ICLR 2020 · 7,418 citations
- Self-Consistency Improves Chain of Thought Reasoning in Language ModelsXuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V. Le et al.ICLR 2023 · 681 citations
- QASC: A Dataset for Question Answering via Sentence CompositionTushar Khot, Peter Clark, Michal Guerquin, Peter Jansen et al.AAAI 2020 · 387 citations
- Deductive Verification of Chain-of-Thought ReasoningZhan Ling, Yunhao Fang, Xuanlin Li, Zhiao Huang et al.NeurIPS 2023 · 234 citations
Related papers
- RLET: A Reinforcement Learning Based Approach for Explainable QA with Entailment TreesTengxiao Liu, Qipeng Guo, Xiangkun Hu, Yue Zhang et al.EMNLP 2022 · 8 citations
- Explaining Answers with Entailment TreesBhavana Dalvi, Peter Jansen, Oyvind Tafjord, Zhengnan Xie et al.EMNLP 2021 · 6 citations
- Natural Language Deduction through Search over Statement CompositionsKaj Bostrom, Zayne Sprague, Swarat Chaudhuri, Greg DurrettEMNLP Findings 2022 · 16 citations
- Natural Logic at the Core: Dynamic Rewards for Entailment Tree GenerationJihao Shi, Xiao Ding, Kai Xiong, Hengwei Zhao et al.ACL Findings 2025
- Faithful Question Answering with Monte-Carlo PlanningRuixin Hong, Hongming Zhang, Hong Zhao, Dong Yu et al.ACL 2023 · 7 citations
