ProoFVer: Natural Logic Theorem Proving for Fact Verification

• URL: http://arxiv.org/abs/2108.11357v1
• Date: Wed, 25 Aug 2021 17:23:04 GMT
• Title: ProoFVer: Natural Logic Theorem Proving for Fact Verification
• Authors: Amrith Krishna, Sebastian Riedel, Andreas Vlachos
• Abstract summary: We propose ProoFVer, a proof system for fact verification using natural logic. The generation of proofs makes ProoFVer an explainable system. We find that humans correctly simulate ProoFVer's decisions more often using the proofs.
• Score: 24.61301908217728
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose ProoFVer, a proof system for fact verification using natural logic. The textual entailment model in ProoFVer is a seq2seq model generating valid natural-logic based logical inferences as its proofs. The generation of proofs makes ProoFVer an explainable system. The proof consists of iterative lexical mutations of spans in the claim with spans in a set of retrieved evidence sentences. Further, each such mutation is marked with an entailment relation using natural logic operators. The veracity of a claim is determined solely based on the sequence of natural logic relations present in the proof. By design, this makes ProoFVer a faithful by construction system that generates faithful explanations. ProoFVer outperforms existing fact-verification models, with more than two percent absolute improvements in performance and robustness. In addition to its explanations being faithful, ProoFVer also scores high on rationale extraction, with a five point absolute improvement compared to attention-based rationales in existing models. Finally, we find that humans correctly simulate ProoFVer's decisions more often using the proofs, than the decisions of an existing model that directly use the retrieved evidence for decision making.

Related papers

• TabVer: Tabular Fact Verification with Natural Logic [11.002475880349452]
  We propose a set-theoretic interpretation of numerals and arithmetic functions in the context of natural logic. We leverage large language models to generate arithmetic expressions by generating questions about salient parts of a claim which are answered by executing functions on tables. In a few-shot setting on FEVEROUS, we achieve an accuracy of 71.4, outperforming both fully neural and symbolic reasoning models by 3.4 points.
  arXiv  Detail & Related papers  (2024-11-02T00:36:34Z)
• Lean-STaR: Learning to Interleave Thinking and Proving [53.923617816215774]
  We present Lean-STaR, a framework for training language models to produce informal thoughts prior to each step of a proof. Lean-STaR achieves state-of-the-art results on the miniF2F-test benchmark within the Lean theorem proving environment.
  arXiv  Detail & Related papers  (2024-07-14T01:43:07Z)
• QA-NatVer: Question Answering for Natural Logic-based Fact Verification [11.002475880349452]
  We propose to use question answering to predict natural logic operators. In a few-shot setting on FEVER, our approach outperforms the best baseline by $4.3$ accuracy points. A human evaluation indicates that our approach produces more plausible with fewer erroneous natural logic operators than previous natural logic-based systems.
  arXiv  Detail & Related papers  (2023-10-22T06:27:31Z)
• Logical Satisfiability of Counterfactuals for Faithful Explanations in NLI [60.142926537264714]
  We introduce the methodology of Faithfulness-through-Counterfactuals. It generates a counterfactual hypothesis based on the logical predicates expressed in the explanation. It then evaluates if the model's prediction on the counterfactual is consistent with that expressed logic.
  arXiv  Detail & Related papers  (2022-05-25T03:40:59Z)
• Generating Natural Language Proofs with Verifier-Guided Search [74.9614610172561]
  We present a novel stepwise method NLProofS (Natural Language Proof Search) NLProofS learns to generate relevant steps conditioning on the hypothesis. It achieves state-of-the-art performance on EntailmentBank and RuleTaker.
  arXiv  Detail & Related papers  (2022-05-25T02:22:30Z)
• FaiRR: Faithful and Robust Deductive Reasoning over Natural Language [25.319674132967553]
  We frame the deductive logical reasoning task by defining three modular components: rule selection, fact selection, and knowledge composition. We observe that FaiRR is robust to novel language perturbations, and is faster at inference than previous works on existing reasoning datasets.
  arXiv  Detail & Related papers  (2022-03-19T07:18:13Z)
• multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning [73.09791959325204]
  We focus on a type of linguistic formal reasoning where the goal is to reason over explicit knowledge in the form of natural language facts and rules. A recent work, named PRover, performs such reasoning by answering a question and also generating a proof graph that explains the answer. In our work, we address a new and challenging problem of generating multiple proof graphs for reasoning over natural language rule-bases.
  arXiv  Detail & Related papers  (2021-06-02T17:58:35Z)
• AmbiFC: Fact-Checking Ambiguous Claims with Evidence [57.7091560922174]
  We present AmbiFC, a fact-checking dataset with 10k claims derived from real-world information needs. We analyze disagreements arising from ambiguity when comparing claims against evidence in AmbiFC. We develop models for predicting veracity handling this ambiguity via soft labels.
  arXiv  Detail & Related papers  (2021-04-01T17:40:08Z)
• ProofWriter: Generating Implications, Proofs, and Abductive Statements over Natural Language [19.917022148887273]
  Transformers have been shown to emulate logical deduction over natural language theories. We show that a generative model, called ProofWriter, can reliably generate both implications of a theory and the natural language proof(s) that support them.
  arXiv  Detail & Related papers  (2020-12-24T00:55:46Z)
• Measuring Association Between Labels and Free-Text Rationales [60.58672852655487]
  In interpretable NLP, we require faithful rationales that reflect the model's decision-making process for an explained instance. We demonstrate that pipelines, existing models for faithful extractive rationalization on information-extraction style tasks, do not extend as reliably to "reasoning" tasks requiring free-text rationales. We turn to models that jointly predict and rationalize, a class of widely used high-performance models for free-text rationalization whose faithfulness is not yet established.
  arXiv  Detail & Related papers  (2020-10-24T03:40:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.