A Catalog of Data Smells for Coding Tasks

A Vitale, R Oliveto, S Scalabrino - ACM Transactions on Software …, 2024 - dl.acm.org
Large Language Models (LLMs) are increasingly becoming fundamental in supporting
software developers in coding tasks. The massive datasets used for training LLMs are often …

On the evaluation of large language models in unit test generation

L Yang, C Yang, S Gao, W Wang, B Wang… - Proceedings of the 39th …, 2024 - dl.acm.org
Unit testing is an essential activity in software development for verifying the correctness of
software components. However, manually writing unit tests is challenging and time …

[PDF][PDF] An empirical study of unit test generation with large language models

L Yang, C Yang, S Gao, W Wang, B Wang… - arXiv preprint arXiv …, 2024 - eecs.umich.edu
Unit testing is an essential activity in software development for verifying the correctness of
software components. However, manually writing unit tests is challenging and time …

Calico: Automated Knowledge Calibration and Diagnosis for Elevating AI Mastery in Code Tasks

Y Qiu, J Hu, Q Zhang, H Yin - Proceedings of the 33rd ACM SIGSOFT …, 2024 - dl.acm.org
Recent advancements in large language models (LLMs) have exhibited promising
capabilities in addressing various tasks such as defect detection and program repair …

Split and Merge: Aligning Position Biases in LLM-based Evaluators

Z Li, C Wang, P Ma, D Wu, S Wang… - Proceedings of the …, 2024 - aclanthology.org
Large language models (LLMs) have shown promise as automated evaluators for assessing
the quality of answers generated by AI systems. However, LLM-based evaluators exhibit …

Efficient training and inference: Techniques for large language models using llama

SR Cunningham, D Archambault, A Kung - Authorea Preprints, 2024 - techrxiv.org
To enhance the efficiency of language models, it would involve optimizing their training and
inference processes to reduce computational demands while maintaining high performance …

Inversion-guided Defense: Detecting Model Stealing Attacks by Output Inverting

S Zhou, T Zhu, D Ye, W Zhou… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Model stealing attacks involve creating copies of machine learning models that have similar
functionalities to the original model without proper authorization. Such attacks raise …

Llms can defend themselves against jailbreaking in a practical manner: A vision paper

D Wu, S Wang, Y Liu, N Liu - arXiv preprint arXiv:2402.15727, 2024 - arxiv.org
Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed
in off-the-shelf large language models (LLMs). A considerable amount of research exists …

Enhancing the resilience of llms against grey-box extractions

H Huang, Y Li, B Jiang, B Jiang, L Liu, Z Liu… - ICML 2024 Next …, 2024 - openreview.net
Large language models are deployed as either closed-source, providing superior
performance with limited customization, or open-source, ensuring full transparency at the …

Transformers: A Security Perspective

BS Latibari, N Nazari, MA Chowdhury, KI Gubbi… - IEEE …, 2024 - ieeexplore.ieee.org
The Transformers architecture has recently emerged as a revolutionary paradigm in the field
of deep learning, particularly excelling in Natural Language Processing (NLP) and …