Do you want to publish a course? Click here

multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning

multiprover: توليد البراهين المتعددة لتحسين الترجمة الشفوية في القاعدة

382   0   0   0.0 ( 0 )
 Publication date 2021
and research's language is English
 Created by Shamra Editor




Ask ChatGPT about the research

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 (Clark et al., 2020). A recent work, named PRover (Saha et al., 2020), performs such reasoning by answering a question and also generating a proof graph that explains the answer. However, compositional reasoning is not always unique and there may be multiple ways of reaching the correct answer. Thus, in our work, we address a new and challenging problem of generating multiple proof graphs for reasoning over natural language rule-bases. Each proof provides a different rationale for the answer, thereby improving the interpretability of such reasoning systems. In order to jointly learn from all proof graphs and exploit the correlations between multiple proofs for a question, we pose this task as a set generation problem over structured output spaces where each proof is represented as a directed graph. We propose two variants of a proof-set generation model, multiPRover. Our first model, Multilabel-multiPRover, generates a set of proofs via multi-label classification and implicit conditioning between the proofs; while the second model, Iterative-multiPRover, generates proofs iteratively by explicitly conditioning on the previously generated proofs. Experiments on multiple synthetic, zero-shot, and human-paraphrased datasets reveal that both multiPRover models significantly outperform PRover on datasets containing multiple gold proofs. Iterative-multiPRover obtains state-of-the-art proof F1 in zero-shot scenarios where all examples have single correct proofs. It also generalizes better to questions requiring higher depths of reasoning where multiple proofs are more frequent.

References used
https://aclanthology.org/
rate research

Read More

The understanding of time expressions includes two sub-tasks: recognition and normalization. In recent years, significant progress has been made in the recognition of time expressions while research on normalization has lagged behind. Existing SOTA n ormalization methods highly rely on rules or grammars designed by experts, which limits their performance on emerging corpora, such as social media texts. In this paper, we model time expression normalization as a sequence of operations to construct the normalized temporal value, and we present a novel method called ARTime, which can automatically generate normalization rules from training data without expert interventions. Specifically, ARTime automatically captures possible operation sequences from annotated data and generates normalization rules on time expressions with common surface forms. The experimental results show that ARTime can significantly surpass SOTA methods on the Tweets benchmark, and achieves competitive results with existing expert-engineered rule methods on the TempEval-3 benchmark.
Despite the recent successes of transformer-based models in terms of effectiveness on a variety of tasks, their decisions often remain opaque to humans. Explanations are particularly important for tasks like offensive language or toxicity detection o n social media because a manual appeal process is often in place to dispute automatically flagged content. In this work, we propose a technique to improve the interpretability of these models, based on a simple and powerful assumption: a post is at least as toxic as its most toxic span. We incorporate this assumption into transformer models by scoring a post based on the maximum toxicity of its spans and augmenting the training process to identify correct spans. We find this approach effective and can produce explanations that exceed the quality of those provided by Logistic Regression analysis (often regarded as a highly-interpretable model), according to a human study.
Grammatical rules are deduced from Arabic spoken by ideally intuitive Arabic speakers, and illustration is the spirit of the rule, endowing it with life, pleasure, and originality. The Arabic used in illustration is that of the holy Quran, sayings of the Prophet as well as renowned poetic and prosaic statements by Arabs from the Jahileah period up to 150 Hizra,i.e, the end of the period of providing arguments. The term illustration is an original Arabic term that came out of Arab concern over mistakes in Arabic. The holy Quran is the source of illustrations, as it is the pillar upon which all other illustrations depend. This paper tries to study the relationship between the grammatical rule and illustrations as well as to demonstrate the motives for illustration, its mechanism, principles, and sources. It also tries to address some equivalents such as provision of argument and evidence as well as analogy.
One of the principal applications of fuzzy logic is in control system design. Fuzzy logic controllers (FLC) can be used to control systems where the use of conventional control techniques may be Problematic. The tuning of fuzzy controllers has tende d to rely on human expert knowledge, but where the number of rules and fuzzy sets is large. The Problem of generation desirable fuzzy rule is very important in the development of fuzzy systems. The purpose of this paper is to present a generation method of fuzzy control rules by learning from examples using genetic algorithms (GA). We propose real coded genetic algorithms (RCGA) for learning fuzzy rules, and an iterative process for obtaining set of rules which covers the examples set with a covering value previously defined.
Pre-trained language models perform well on a variety of linguistic tasks that require symbolic reasoning, raising the question of whether such models implicitly represent abstract symbols and rules. We investigate this question using the case study of BERT's performance on English subject--verb agreement. Unlike prior work, we train multiple instances of BERT from scratch, allowing us to perform a series of controlled interventions at pre-training time. We show that BERT often generalizes well to subject--verb pairs that never occurred in training, suggesting a degree of rule-governed behavior. We also find, however, that performance is heavily influenced by word frequency, with experiments showing that both the absolute frequency of a verb form, as well as the frequency relative to the alternate inflection, are causally implicated in the predictions BERT makes at inference time. Closer analysis of these frequency effects reveals that BERT's behavior is consistent with a system that correctly applies the SVA rule in general but struggles to overcome strong training priors and to estimate agreement features (singular vs. plural) on infrequent lexical items.

suggested questions

comments
Fetching comments Fetching comments
Sign in to be able to follow your search criteria
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا