English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Logical reasoning is fundamental for humans yet presents a substantialchallenge in the domain of Artificial Intelligence. Initially, researchers usedKnowledge Representation and Reasoning (KR) systems that did not scale andrequired non trivial manual effort. Recently, the emergence of large languagemodels (LLMs) has demonstrated the ability to overcome various limitations offormal Knowledge Representation (KR) systems. Consequently, there is a growinginterest in using LLMs for logical reasoning via natural language. This workstrives to understand the proficiency of LLMs in logical reasoning by offeringa brief review of the latest progress in this area; with a focus on the logicalreasoning datasets, tasks, and the methods adopted to utilize LLMs forreasoning. To offer a thorough analysis, we have compiled a benchmark titledLogiGLUE. This includes 24 varied datasets encompassing deductive, abductive,and inductive reasoning. We have standardized these datasets into Seq2Seq tasksto facilitate straightforward training and evaluation for future research.Utilizing LogiGLUE as a foundation, we have trained an instruction fine tunedlanguage model, resulting in LogiT5. We study single task training, multi tasktraining, and a chain of thought knowledge distillation fine tuning techniqueto assess the performance of model across the different logical reasoningcategories. By this comprehensive process, we aim to shed light on thecapabilities and potential pathways for enhancing logical reasoning proficiencyin LLMs, paving the way for more advanced and nuanced developments in thiscritical field.

Towards LogiGLUE: A Brief Survey and A Benchmark for Analyzing Logical Reasoning Capabilities of Language Models