An Investigation into the Utility of Large Language Models in Geotechnical Education and Problem Solving

Chen, Liuxin; Tophel, Amir; Hettiyadura, Umidu; Kodikara, Jayantha

doi:10.3390/geotechnics4020026

Cited by 4 publications

(2 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…LLMs could potentially address these challenges by offering personalised learning experiences and adaptive content delivery, thus enhancing educational outcomes (Yang et al, 2024). Yet, the implementation of LLMs in this field faces significant hurdles, particularly in their ability to accurately integrate and apply domain-specific formulas and concepts, which are essential for solving specialised problems (L. Chen et al, 2024).…”

Section: Introductionmentioning

confidence: 99%

“…The authors' previous study showed how the efficiency of GPT-4 can be improved by incorporating Chain of Thought (CoT) and our proposed custom instruction techniques. The authors showed that by utilising these two techniques, the accuracy of GPT-4 was improved from 29% to 34% and 67% for CoT and custom instruction, respectively (L. Chen et al, 2024). Despite these improvements, other development methods may offer unique benefits that enhance the effectiveness and reliability of AI-driven educational tools.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards an AI Tutor for Undergraduate Geotechnical Engineering: A Comparative Study of Evaluating the Efficiency of Large Language Model Application Programming Interfaces

Tophel,

Chen,

Hettiyadura

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

This study investigates the efficiency of Large Language Model (LLM) Application Programming Interfaces (APIs)—specifically GPT-4 and Llama-3—as AI tutors for undergraduate Geotechnical Engineering education. As educational needs in specialised fields like Geotechnical Engineering become increasingly complex, innovative teaching tools that provide personalised learning experiences are essential. This research evaluates the capabilities of GPT-4’s and Llama-3’s APIs in integrating and applying Geotechnical Engineering formulas, offering accurate problem-solving and explanatory responses, and adapting to varied educational requirements. Using comparative analysis, the study employs a formula integration approach known as Retrieval-Augmented Generation (RAG) with two widely used LLM models, GPT-4 and Llama-3. A set of 20 challenging questions, previously identified as problematic for zero-shot solutions for GPT-4, served as the evaluation basis. The models were assessed on accuracy, formula integration, clarity of explanation, and problem-solving adaptability. Results indicate that GPT-4 and Llama-3 have significant potential as AI tutors in Geotechnical Engineering. GPT-4, utilising RAG, demonstrated superior performance, correctly answering 95% of the questions at a temperature setting of 0.1, 82.5% at 0.5, and 60% at 1. In contrast, Llama-3 correctly answered 25% of the questions in zero-shot tasks and 45% with API by setting a temperature of 0.1. The study underscores the need for advanced formula integration techniques and domain-specific training to enhance the educational utility of LLM APIs. Future research should focus on refining formula integration methods, expanding domain-specific knowledge bases, and assessing long-term learning outcomes. This work contributes to the ongoing dialogue on AI in education, providing insights into deploying LLMs as personalised, effective teaching aids in engineering disciplines.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Towards an AI Tutor for Undergraduate Geotechnical Engineering: A Comparative Study of Evaluating the Efficiency of Large Language Model Application Programming Interfaces

Tophel,

Chen,

Hettiyadura

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Leveraging Large Language Models for Enhancing Literature-Based Discovery

Taleb,

Navaz,

Serhani

2024

BDCC

View full text Add to dashboard Cite

The exponential growth of biomedical literature necessitates advanced methods for Literature-Based Discovery (LBD) to uncover hidden, meaningful relationships and generate novel hypotheses. This research integrates Large Language Models (LLMs), particularly transformer-based models, to enhance LBD processes. Leveraging LLMs’ capabilities in natural language understanding, information extraction, and hypothesis generation, we propose a framework that improves the scalability and precision of traditional LBD methods. Our approach integrates LLMs with semantic enhancement tools, continuous learning, domain-specific fine-tuning, and robust data cleansing processes, enabling automated analysis of vast text and identification of subtle patterns. Empirical validations, including scenarios on the effects of garlic on blood pressure and nutritional supplements on health outcomes, demonstrate the effectiveness of our LLM-based LBD framework in generating testable hypotheses. This research advances LBD methodologies, fosters interdisciplinary research, and accelerates discovery in the biomedical domain. Additionally, we discuss the potential of LLMs in drug discovery, highlighting their ability to extract and present key information from the literature. Detailed comparisons with traditional methods, including Swanson’s ABC model, highlight our approach’s advantages. This comprehensive approach opens new avenues for knowledge discovery and has the potential to revolutionize research practices. Future work will refine LLM techniques, explore Retrieval-Augmented Generation (RAG), and expand the framework to other domains, with a focus on dehallucination.

show abstract

TransSMPL: Efficient Human Pose Estimation with Pruned and Quantized Transformer Networks

Kim,

Yoo,

Ryu

et al. 2024

Electronics

View full text Add to dashboard Cite

Existing Transformers for 3D human pose and shape estimation models often struggle with computational complexity, particularly when handling high-resolution feature maps. These challenges limit their ability to efficiently utilize fine-grained features, leading to suboptimal performance in accurate body reconstruction. In this work, we propose TransSMPL, a novel Transformer framework built upon the SMPL model, specifically designed to address the challenges of computational complexity and inefficient utilization of high-resolution feature maps in 3D human pose and shape estimation. By replacing HRNet with MobileNetV3 for lightweight feature extraction, applying pruning and quantization techniques, and incorporating an early exit mechanism, TransSMPL significantly reduces both computational cost and memory usage. TransSMPL introduces two key innovations: (1) a multi-scale attention mechanism, reduced from four scales to two, allowing for more efficient global and local feature integration, and (2) a confidence-based early exit strategy, which enables the model to halt further computations when high-confidence predictions are achieved, further enhancing efficiency. Extensive pruning and dynamic quantization are also applied to reduce the model size while maintaining competitive performance. Quantitative and qualitative experiments on the Human3.6M dataset demonstrate the efficacy of TransSMPL. Our model achieves an MPJPE (Mean Per Joint Position Error) of 48.5 mm, reducing the model size by over 16% compared to existing methods while maintaining a similar level of accuracy.

show abstract

An Investigation into the Utility of Large Language Models in Geotechnical Education and Problem Solving

Cited by 4 publications

References 37 publications

Towards an AI Tutor for Undergraduate Geotechnical Engineering: A Comparative Study of Evaluating the Efficiency of Large Language Model Application Programming Interfaces

Towards an AI Tutor for Undergraduate Geotechnical Engineering: A Comparative Study of Evaluating the Efficiency of Large Language Model Application Programming Interfaces

Leveraging Large Language Models for Enhancing Literature-Based Discovery

TransSMPL: Efficient Human Pose Estimation with Pruned and Quantized Transformer Networks

Contact Info

Product

Resources

About