Search

Large language models (LLMs) have achieved significant performance in various natural language reasoning tasks. However, they still struggle with performing first-order logic reasoning over formal logical theories expressed in natural language. This is because the previous LLMs-based reasoning systems have the theoretical incompleteness issue. As a result, it can only address a limited set of simple reasoning problems, which significantly decreases their generalization ability. To address this issue, we propose a novel framework, named Generalizable and Faithful Reasoner (GFaiR), which introduces the paradigm of resolution refutation. Resolution refutation has the capability to solve all first-order logic reasoning problems by extending reasoning rules and employing the principle of proof by contradiction, so our system's completeness can be improved by introducing resolution refutation. Experimental results demonstrate that our system outperforms previous works by achieving state-of-the-art performances in complex scenarios while maintaining performances in simple scenarios. Besides, we observe that GFaiR is faithful to its reasoning process., Comment: LREC-Coling 2024

Causal reasoning ability is crucial for numerous NLP applications. Despite the impressive emerging ability of ChatGPT in various NLP tasks, it is unclear how well ChatGPT performs in causal reasoning. In this paper, we conduct the first comprehensive evaluation of the ChatGPT's causal reasoning capabilities. Experiments show that ChatGPT is not a good causal reasoner, but a good causal explainer. Besides, ChatGPT has a serious hallucination on causal reasoning, possibly due to the reporting biases between causal and non-causal relationships in natural language, as well as ChatGPT's upgrading processes, such as RLHF. The In-Context Learning (ICL) and Chain-of-Thought (CoT) techniques can further exacerbate such causal hallucination. Additionally, the causal reasoning ability of ChatGPT is sensitive to the words used to express the causal concept in prompts, and close-ended prompts perform better than open-ended prompts. For events in sentences, ChatGPT excels at capturing explicit causality rather than implicit causality, and performs better in sentences with lower event density and smaller lexical distance between events. The code is available on https://github.com/ArrogantL/ChatGPT4CausalReasoning ., Comment: Accepted to Findings of EMNLP 2023

Given data with label noise (i.e., incorrect data), deep neural networks would gradually memorize the label noise and impair model performance. To relieve this issue, curriculum learning is proposed to improve model performance and generalization by ordering training samples in a meaningful (e.g., easy to hard) sequence. Previous work takes incorrect samples as generic hard ones without discriminating between hard samples (i.e., hard samples in correct data) and incorrect samples. Indeed, a model should learn from hard samples to promote generalization rather than overfit to incorrect ones. In this paper, we address this problem by appending a novel loss function DiscrimLoss, on top of the existing task loss. Its main effect is to automatically and stably estimate the importance of easy samples and difficult samples (including hard and incorrect samples) at the early stages of training to improve the model performance. Then, during the following stages, DiscrimLoss is dedicated to discriminating between hard and incorrect samples to improve the model generalization. Such a training strategy can be formulated dynamically in a self-supervised manner, effectively mimicking the main principle of curriculum learning. Experiments on image classification, image regression, text sequence regression, and event relation reasoning demonstrate the versatility and effectiveness of our method, particularly in the presence of diversified noise levels.

A key design parameter for high-density, thickened tailings (paste) is the yield stress, widely measured by the quick and easy slump test. The paper investigates current techniques for interpreting the cylindrical slump (or mini-slump) test, taking into account the lifting of the cylindrical mould in numerical simulations using a computational fluid dynamics approach. Simulations with different mould lifting velocities were carried out. The influence of plastic viscosity and yield stress on mini-slump test results was first studied using a mould lifting velocity of 0.01 m/s, representative of rates used in laboratory tests. To understand the influence of mould lifting velocity, the predicted slump and spread from mini-slump test simulations for three different scenarios (v(lifting) 0.002 m/s, v(lifting) 0.01 m/s and no lifting i.e. mould disappears instantaneously) were compared with those from vane viscometer experiments on the rheological properties of kaolin. The results suggest that the lifting speed of the mould has a significant influence on the mini-slump test result, which must therefore be taken into account in both numerical simulations and laboratory tests. It was found that the variation of mould lifting velocity had a greater influence on slump than spread, indicating that spread is a more appropriate measurement for determining the yield stress in a mini-slump test. This was particularly true for relatively low yield stresses of 60 Pa or less, which are values typical of most thickened tailings deposits currently operating internationally., A key design parameter for high-density, thickened tailings (paste) is the yield stress, widely measured by the quick and easy slump test. The paper investigates current techniques for interpreting the cylindrical slump (or mini-slump) test, taking into account the lifting of the cylindrical mould in numerical simulations using a computational fluid dynamics approach. Simulations with different mould lifting velocities were carried out. The influence of plastic viscosity and yield stress on mini-slump test results was first studied using a mould lifting velocity of 0.01 m/s, representative of rates used in laboratory tests. To understand the influence of mould lifting velocity, the predicted slump and spread from mini-slump test simulations for three different scenarios (v(lifting) 0.002 m/s, v(lifting) 0.01 m/s and no lifting i.e. mould disappears instantaneously) were compared with those from vane viscometer experiments on the rheological properties of kaolin. The results suggest that the lifting speed of the mould has a significant influence on the mini-slump test result, which must therefore be taken into account in both numerical simulations and laboratory tests. It was found that the variation of mould lifting velocity had a greater influence on slump than spread, indicating that spread is a more appropriate measurement for determining the yield stress in a mini-slump test. This was particularly true for relatively low yield stresses of 60 Pa or less, which are values typical of most thickened tailings deposits currently operating internationally.