The deductive closure of an ideal knowledge base (KB) contains exactly the logical queries that the KB can answer. However, in practice KBs are both incomplete and over-specified, failing to answer some queries that have real-world answers. \emph{Query embedding} (QE) techniques have been recently proposed where KB entities and KB queries are represented jointly in an embedding space, supporting relaxation and generalization in KB inference. However, experiments in this paper show that QE systems may disagree with deductive reasoning on answers that do not require generalization or relaxation. We address this problem with a novel QE method that is more faithful to deductive reasoning, and show that this leads to better performance on complex queries to incomplete KBs. Finally we show that inserting this new QE module into a neural question-answering system leads to substantial improvements over the state-of-the-art.
Teaching Language Models to Faithfully Express their Uncertainty
University of Amsterdam（阿姆斯特丹大学）；Outsystems（Outsystems公司）；Instituto de Telecomunicações（电信研究所）；Instituto Superior Técnico, Universidade de Lisboa（里斯本大学电信研究所）
FAIR-RAG: Faithful Adaptive Iterative Refinement for Retrieval-Augmented Generation
Sharif University of Technology（谢里夫理工大学）；Iran University of Science and Technology（伊朗科学技术大学）；Noor Avaran Jelvehaye Maanaei Najm Co., Ltd.（诺尔·阿瓦兰·杰拉维耶·马纳埃尼纳姆公司）
MURR: Model Updating with Regularized Replay for Searching a Document Stream
Johns Hopkins University（约翰斯·霍普金斯大学）；University of Pisa（比萨大学）；University of Glasgow（格拉斯哥大学）；University of Maryland（马里兰大学）；Information Sciences Institute, University of Southern California（南加州大学信息科学研究所）
Cornell University（康奈尔大学）；New York University（纽约大学）；The George Washington University（乔治华盛顿大学）；Amazon Web Services AI（亚马逊网络服务人工智能）
Neural abstractive summarization models are prone to generate content inconsistent with the source document, i.e. unfaithful. Existing automatic metrics do not capture such mistakes effectively. We tackle the problem of evaluating faithfulness of a generated summary given its source document. We first collected human annotations of faithfulness for outputs from numerous models on two datasets. We find that current models exhibit a trade-off between abstractiveness and faithfulness: outputs with less word overlap with the source document are more likely to be unfaithful. Next, we propose an aut…
One of the fundamental problems in Artificial Intelligence is to perform complex multi-hop logical reasoning over the facts captured by a knowledge graph (KG). This problem is challenging, because KGs can be massive and incomplete. Recent approaches embed KG entities in a low dimensional space and then use these embeddings to find the answer entities. However, it has been an outstanding challenge of how to handle arbitrary first-order logic (FOL) queries as present methods are limited to only a subset of FOL operators. In particular, the negation operator is not supported. An additional limita…
There is broad agreement in the literature that explanation methods should be faithful to the model that they explain, but faithfulness remains a rather vague term. We revisit faithfulness in the context of continuous data and propose two formal definitions of faithfulness for feature attribution methods. Qualitative faithfulness demands that scores reflect the true qualitative effect (positive vs. negative) of the feature on the model and quanitative faithfulness that the magnitude of scores reflect the true quantitative effect. We discuss under which conditions these requirements can be sati…
