Showing papers on "Adaptive reasoning published in 2022"
TL;DR: In this paper , the authors describe the spatial reasoning of MTs students in imagining and explaining parts of objects, mentioning and describing parts of object and changing the shape of objects from the problems given.
Abstract: Spatial reasoning as an important component in students' mathematical thinking and problem solving. How do students think spatially, how is the flow in solving contextual problems, this is spatial reasoning. Spatial reasoning consists of three characteristics, namely perceiving, visual-spatial and transforming. This research is included in descriptive qualitative which aims to describe the spatial reasoning of MTs students in imagining and explaining parts of objects, mentioning and describing parts of objects and changing the shape of objects from the problems given. The subjects in this study were three students from class VIII of MTs Negeri 3 Muara Ampolu Tapanuli Selatan who were taken from each level of mathematical ability, namely high, medium and low abilities after being given an initial math ability test. To find out the profile of students' spatial reasoning in contextual solving, the research subjects were given a spatial reasoning test. The results showed that students with high and moderate mathematical abilities in solving contextual problems that required spatial reasoning had good results. Meanwhile, low-ability students have difficulty in solving contextual problems so that their spatial reasoning has poor results.
20 Oct 2022
TL;DR: In this paper , a unified reasoning framework for mirroring the modular and compositional reasoning process of humans in complex decision-making is presented. But the framework is limited to solving multiple tasks with a single model and is trained and inferred in an end-to-end manner.
Abstract: This paper presents ReasonFormer, a unified reasoning framework for mirroring the modular and compositional reasoning process of humans in complex decision-making. Inspired by dual-process theory in cognitive science, the representation module (automatic thinking) and reasoning modules (controlled thinking) are decoupled to capture different levels of cognition. Upon the top of the representation module, the pre-trained reasoning modules are modular and professional in specific and fundamental reasoning skills (e.g., logic, simple QA, etc). To mimic the controlled compositional thinking process, different reasoning modules are dynamically activated and composed in both parallel and cascaded manners to control what reasoning skills are activated and how deep the reasoning process will be reached to solve the current problems. The unified reasoning framework solves multiple tasks with a single model, and is trained and inferred in an end-to-end manner. Evaluated on 11 datasets requiring different reasoning skills and complexity, ReasonFormer demonstrates substantial performance boosts, revealing the compositional reasoning ability. Few-shot experiments exhibit better generalization ability by learning to compose pre-trained skills for new tasks with limited data, and decoupling the representation module and the reasoning modules. Further analysis shows the modularity of reasoning modules as different tasks activate distinct reasoning skills at different reasoning depths.
28 Oct 2022
TL;DR: CoRe as discussed by the authors proposes a cooperative reasoning-induced pre-trained language model for solving the math word problem (MWP), where the generator is responsible for generating reasoning paths, and the verifiers are used to supervise the evaluation in order to obtain reliable feedback for the generator.
Abstract: Large-scale pre-trained language models (PLMs) bring new opportunities to challenging problems, especially those that need high-level intelligence, such as the math word problem (MWPs). However, directly applying existing PLMs to MWPs can fail as the generation process lacks sufficient supervision and thus lacks fast adaptivity as humans. We notice that human reasoning has a dual reasoning framework that consists of an immediate reaction system (system 1) and a delicate reasoning system (system 2), where the entire reasoning is determined by their interaction. This inspires us to develop a cooperative reasoning-induced PLM for solving MWPs, called Cooperative Reasoning (CoRe), resulting in a human-like reasoning architecture with system 1 as the generator and system 2 as the verifier. In our approach, the generator is responsible for generating reasoning paths, and the verifiers are used to supervise the evaluation in order to obtain reliable feedback for the generator. We evaluate our CoRe framework on several mathematical reasoning datasets and achieve decent improvement over state-of-the-art methods, up to 9.6% increase over best baselines. Our codes are available at https://github.com/TianHongZXY/CoRe