Reasoning is key to many decision making processes. It requires consolidating a set of rule-like premises that are often associated with degrees of uncertainty and observations to draw conclusions. In this work, we address both the case where premises are specified as numeric probabilistic rules and situations in which humans state their estimates using words expressing degrees of certainty. Existing probabilistic reasoning datasets simplify the task, e.g., by requiring the model to only rank textual alternatives, by including only binary random variables, or by making use of a limited set of templates that result in less varied text.In this work, we present QUITE, a question answering dataset of real-world Bayesian reasoning scenarios with categorical random variables and complex relationships. QUITE provides high-quality natural language verbalizations of premises together with evidence statements and expects the answer to a question in the form of an estimated probability. We conduct an extensive set of experiments, finding that logic-based models outperform out-of-the-box large language models on all reasoning types (causal, evidential, and explaining-away). Our results provide evidence that neuro-symbolic models are a promising direction for improving complex reasoning. We release QUITE and code for training and experiments on Github.

Structured knowledge bases (KBs) are an asset for search engines and other applications but are inevitably incomplete. Language models (LMs) have been proposed for unsupervised knowledge base completion (KBC), yet, their ability to do this at scale and with high accuracy remains an open question. Prior experimental studies mostly fall short because they only evaluate on popular subjects, or sample already existing facts from KBs. In this work, we perform a careful evaluation of GPT’s potential to complete the largest public KB: Wikidata. We find that, despite their size and capabilities, models like GPT-3, ChatGPT and GPT-4 do not achieve fully convincing results on this task. Nonetheless, it provides solid improvements over earlier approaches with smaller LMs. In particular, we show that it is feasible to extend Wikidata by 27M facts at 90% precision.

The widespread usage of latent language representations via pre-trained language models (LMs) suggests that they are a promising source of structured knowledge. However, existing methods focus only on a single object per subject-relation pair, even though often multiple objects are correct. To overcome this limitation, we analyze these representations for their potential to yield materialized multi-object relational knowledge. We formulate the problem as a rank-then-select task. For ranking candidate objects, we evaluate existing prompting techniques and propose new ones incorporating domain knowledge. Among the selection methods, we find that choosing objects with a likelihood above a learned relation-specific threshold gives a 49.5% F1 score. Our results highlight the difficulty of employing LMs for the multi-valued slot-filling task, and pave the way for further research on extracting relational knowledge from latent language representations.

Despite their impressive scale, knowledge bases (KBs), such as Wikidata, still contain significant gaps. Language models (LMs) have been proposed as a source for filling these gaps. However, prior works have focused on prominent entities with rich coverage by LMs, neglecting the crucial case of long-tail entities. In this paper, we present a novel method for LM-based-KB completion that is specifically geared for facts about long-tail entities. The method leverages two different LMs in two stages: for candidate retrieval and for candidate verification and disambiguation. To evaluate our method and various baselines, we introduce a novel dataset, called MALT, rooted in Wikidata. Our method outperforms all baselines in F1, with major gains especially in recall.

Understanding causality is a core aspect of intelligence. The Event Causality Identification with Causal News Corpus Shared Task addresses two aspects of this challenge: Subtask 1 aims at detecting causal relationships in texts, and Subtask 2 requires identifying signal words and the spans that refer to the cause or effect, respectively. Our system, which is based on pre-trained transformers, stacked sequence tagging, and synthetic data augmentation, ranks third in Subtask 1 and wins Subtask 2 with an F1 score of 72.8, corresponding to a margin of 13 pp. to the second-best system.

Compiling comprehensive repositories of commonsense knowledge is a long-standing problem in AI. Many concerns revolve around the issue of reporting bias, i.e., that frequency in text sources is not a good proxy for relevance or truth. This paper explores whether children’s texts hold the key to commonsense knowledge compilation, based on the hypothesis that such content makes fewer assumptions on the reader’s knowledge, and therefore spells out commonsense more explicitly. An analysis with several corpora shows that children’s texts indeed contain much more, and more typical commonsense assertions. Moreover, experiments show that this advantage can be leveraged in popular language-model-based commonsense knowledge extraction settings, where task-unspecific fine-tuning on small amounts of children texts (childBERT) already yields significant improvements. This provides a refreshing perspective different from the common trend of deriving progress from ever larger models and corpora.

This paper presents a new task of predicting the coverage of a text document for relation extraction (RE): Does the document contain many relational tuples for a given entity? Coverage predictions are useful in selecting the best documents for knowledge base construction with large input corpora. To study this problem, we present a dataset of 31,366 diverse documents for 520 entities. We analyze the correlation of document coverage with features like length, entity mention frequency, Alexa rank, language complexity, and information retrieval scores. Each of these features has only moderate predictive power. We employ methods combining features with statistical models like TF-IDF and language models like BERT. The model combining features and BERT, HERB, achieves an F1 score of up to 46%. We demonstrate the utility of coverage predictions on two use cases: KB construction and claim refutation.

Starting from the COMET methodology by Bosselut et al. (2019), generating commonsense knowledge directly from pre-trained language models has recently received significant attention. Surprisingly, up to now no materialized resource of commonsense knowledge generated this way is publicly available. This paper fills this gap, and uses the materialized resources to perform a detailed analysis of the potential of this approach in terms of precision and recall. Furthermore, we identify common problem cases, and outline use cases enabled by materialized resources. We posit that the availability of these resources is important for the advancement of the field, as it enables an off-the-shelf-use of the resulting knowledge, as well as further analyses on its strengths and weaknesses.

As climate change alters the physical world we inhabit, opinions surrounding this hot-button issue continue to fluctuate. This is apparent on social media, particularly Twitter. In this paper, we explore concrete climate change data concerning the Air Quality Index (AQI), and its relationship to tweets. We incorporate commonsense connotations for appeal to the masses. Earlier work focuses primarily on accuracy and performance of sentiment analysis tools / models, much geared towards experts. We present commonsense interpretations of results, such that they are not impervious to the masses. Moreover, our study uses real data on multiple environmental quantities comprising AQI. We address human sentiments gathered from linked data on hashtagged tweets with geolocations. Tweets are analyzed using VADER, subtly entailing commonsense reasoning. Interestingly, correlations between climate change tweets and air quality data vary not only based upon the year, but also the specific environmental quantity. It is hoped that this study will shed light on possible areas to increase awareness of climate change, and methods to address it, by the scientists as well as the common public. In line with Linked Data initiatives, we aim to make this work openly accessible on a network, published with the Creative Commons license.

Modern web content - news articles, blog posts, educational resources, marketing brochures - is predominantly multimodal. A notable trait is the inclusion of media such as images placed at meaningful locations within a textual narrative. Most often, such images are accompanied by captions - either factual or stylistic (humorous, metaphorical, etc.) - making the narrative more engaging to the reader. While standalone image captioning has been extensively studied, captioning an image based on external knowledge such as its surrounding text remains under-explored. In this paper, we study this new task: given an image and an associated unstructured knowledge snippet, the goal is to generate a contextual caption for the image.

ASCENT is a fully automated methodology for extracting and consolidating commonsense assertions from web contents (Nguyen et al., 2021). It advances traditional triple-based commonsense knowledge representation by capturing semantic facets like locations and purposes, and composite concepts, i.e., subgroups and related aspects of subjects. In this demo, we present a web portal that allows users to understand its construction process, explore its content, and observe its impact in the use case of question answering. The demo website (https://ascent.mpi-inf.mpg.de) and an introductory video (https://youtu.be/qMkJXqu_Yd4) are both available online.

The Internet contains a multitude of social media posts and other of stories where text is interspersed with images. In these contexts, images are not simply used for general illustration, but are judiciously placed in certain spots of a story for multimodal descriptions and narration. In this work we analyze the problem of text-image alignment, and present SANDI, a methodology for automatically selecting images from an image collection and aligning them with text paragraphs of a story. SANDI combines visual tags, user-provided tags and background knowledge, and uses an Integer Linear Program to compute alignments that are semantically meaningful. Experiments show that SANDI can select and align images with texts with high quality of semantic fit.

Fiction and fantasy are archetypes of long-tail domains that lack suitable NLP methodologies and tools. We present ENTYFI, a web-based system for fine-grained typing of entity mentions in fictional texts. It builds on 205 automatically induced high-quality type systems for popular fictional domains, and provides recommendations towards reference type systems for given input texts. Users can exploit the richness and diversity of these reference type systems for fine-grained supervised typing, in addition, they can choose among and combine four other typing modules: pre-trained real-world models, unsupervised dependency-based typing, knowledge base lookups, and constraint-based candidate consolidation. The demonstrator is available at: https://d5demos.mpi-inf.mpg.de/entyfi.

Scalar implicatures are language features that imply the negation of stronger statements, e.g., “She was married twice” typically implicates that she was not married thrice. In this paper we discuss the importance of scalar implicatures in the context of textual information extraction. We investigate how textual features can be used to predict whether a given text segment mentions all objects standing in a certain relationship with a certain subject. Preliminary results on Wikipedia indicate that this prediction is feasible, and yields informative assessments.

Information extraction (IE) from text has largely focused on relations between individual entities, such as who has won which award. However, some facts are never fully mentioned, and no IE method has perfect recall. Thus, it is beneficial to also tap contents about the cardinalities of these relations, for example, how many awards someone has won. We introduce this novel problem of extracting cardinalities and discusses the specific challenges that set it apart from standard IE. We present a distant supervision method using conditional random fields. A preliminary evaluation results in precision between 3% and 55%, depending on the difficulty of relations.