Detecting Micromanagement During Pair Programming

Ubani, Solomon; Nielsen, Rodney D.

doi:10.1109/ithet50392.2021.9759726

Cited by 4 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gao et al (2023) proposed a novel noise-robust re-weighting framework SunGen to automatically construct high-quality data for zero-shot classification problems. Ubani et al (2023) investigated the use of data obtained from prompting a large generative language model, to generate synthetic training data for few-shot learning. Tang et al (2023) proposed to generate a vast quantity of high-quality synthetic data with labels utilizing ChatGPT and fine-tuning a local model for the downstream task.…”

Section: Black-box Kdmentioning

confidence: 99%

Analysis of the breakage parameters of railway ballast based on the discrete element method

Liu

Dai²,

Wang³

et al. 2022

J. Zhejiang Univ. Sci. A

View full text Add to dashboard Cite

Large language models (LLMs) have demonstrated remarkable capabilities across various NLP tasks. However, their computational costs are prohibitively high. To address this issue, previous research has attempted to distill the knowledge of LLMs into smaller models by generating annotated data. Nonetheless, these works have mainly focused on the direct use of LLMs for text generation and labeling, without fully exploring their potential to comprehend the target task and acquire valuable knowledge. In this paper, we propose EvoKD: Evolving Knowledge Distillation, which leverages the concept of active learning to interactively enhance the process of data generation using large language models, simultaneously improving the task capabilities of small domain model (student model). Different from previous work, we actively analyze the student model's weaknesses, and then synthesize labeled samples based on the analysis. In addition, we provide iterative feedback to the LLMs regarding the student model's performance to continuously construct diversified and challenging samples. Experiments and analysis on different NLP tasks, namely, text classification and named entity recognition show the effectiveness of EvoKD.

show abstract

Section: Black-box Kdmentioning

confidence: 99%

Analysis of the breakage parameters of railway ballast based on the discrete element method

Liu

Dai²,

Wang³

et al. 2022

J. Zhejiang Univ. Sci. A

View full text Add to dashboard Cite

show abstract

“…Nascent work also demonstrates that automatically generated annotations for dialog acts are effective for understanding learning. Recent studies in learning analytics employed NLP techniques to analyze collaborative problem-solving, such as identifying collaborative skills through student speech [51], detecting language patterns in pair programming [60], and classifying interactions in collaborative science tasks [21].…”

Section: Introductionmentioning

confidence: 99%

Combining Dialog Acts and Skill Modeling: What Chat Interactions Enhance Learning Rates During AI-Supported Peer Tutoring?

Borchers,

Yang,

Lin

et al. 2024

Preprint

View full text Add to dashboard Cite

Peer tutoring can improve learning by prompting learners to reflect. To assess whether peer interactions are conducive to learning and provide peer tutoring support accordingly, what tutorial dialog types relate to student learning most? Advancements in collaborative learning analytics allow for merging machine learning-based dialog act classification with cognitive modeling of fine-grained learning processes during problem-solving to illuminate this question. We estimate how much peer-tutored students improve in a collaborative tutoring system for linear equation-solving in K-12 mathematics in relationship to the peer dialog types they engage in. This work establishes a reliable BERT classifier with an accuracy of close to 80\% to classify chat messages during peer tutoring into minimal, facilitative, and constructive, serving as instructional factors. Based on data from 394 students, peer tutor dialog was rare. Only 8\% of tutee problem-solving steps were followed by peer tutor chat messages. Still, facilitative tutor dialog was associated with an increased tutee learning rate. Meanwhile, tutor dialog classified as constructive was associated with lower learning rates. Content analysis suggested that such dialog often reinforced incorrect solutions, gave away answers, or was unrelated to the taught content. Hence, considering problem-solving solution contexts could improve the assessment of peer tutoring dialog. Peer tutors engaging in little dialog could be attributed to the high cognitive demand of learning to tutor while still learning the content they tutor on. Providing peer tutors with instructional support to engage in constructive dialog may improve the tutee's learning.

show abstract

“…Nonetheless, existing approaches are inadequate for compressing LLMs due to their exceptionally high compression ratios. Some prior research (Wang et al, 2022;Dai et al, 2023;Ubani et al, 2023) has suggested utilizing LLMs for data augmentation and knowledge transfer to small-scale models, which allows the latter to demonstrate improved performance on lowresource datasets. However, when tackling more challenging tasks like the SuperGLUE benchmark (Wang et al, 2019a), the limited parameter size of small-scale models becomes a hindrance, preventing them from effectively retaining the knowledge transferred by LLMs.…”

Section: Introductionmentioning

confidence: 99%

Retrieval-based Knowledge Transfer: An Effective Approach for Extreme Large Language Model Compression

Liu,

Wang

et al. 2023

Findings of the Association for Computational Linguistics: EMNLP 2023

View full text Add to dashboard Cite

Large-scale pre-trained language models (LLMs) have demonstrated exceptional performance in various natural language processing (NLP) tasks. However, the massive size of these models poses huge challenges for their deployment in real-world applications. While numerous model compression techniques have been proposed, most of them are not wellsuited for achieving extreme model compression when there is a significant gap in model scale. In this paper, we introduce a novel compression paradigm called Retrieval-based Knowledge Transfer (RetriKT), which effectively transfers the knowledge of LLMs to extremely small-scale models (e.g., 1%). In particular, our approach extracts knowledge from LLMs to construct a knowledge store, from which the small-scale model can retrieve relevant information and leverage it for effective inference. To improve the quality of the model, soft prompt tuning and Proximal Policy Optimization (PPO) reinforcement learning techniques are employed. Extensive experiments are conducted on low-resource tasks from Su-perGLUE and GLUE benchmarks. The results demonstrate that the proposed approach significantly enhances the performance of smallscale models by leveraging the knowledge from LLMs.

show abstract

Detecting Micromanagement During Pair Programming

Cited by 4 publications

References 17 publications

Analysis of the breakage parameters of railway ballast based on the discrete element method

Analysis of the breakage parameters of railway ballast based on the discrete element method

Combining Dialog Acts and Skill Modeling: What Chat Interactions Enhance Learning Rates During AI-Supported Peer Tutoring?

Retrieval-based Knowledge Transfer: An Effective Approach for Extreme Large Language Model Compression

Contact Info

Product

Resources

About