applsci-logo

Journal Browser

Journal Browser

Research Progress on LLMs (Large Language Models): Catastrophic Forgetting, Alignment and Safety and Hallucinations

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: 20 August 2025 | Viewed by 1688

Special Issue Editor


E-Mail Website
Guest Editor
Department of Computer Science and Engineering, Ewha Womans University, Seoul, Republic of Korea
Interests: data mining; databases; artificial intelligence; distributed computing; artificial neural network; computing in mathematics, natural science, engineering and medicine; software Engineering

Special Issue Information

Dear Colleagues,

The journal invites submissions to a Special Issue on three issues of LLMs (large language models), as follows:

  1. Forgetting problems in deep neural networks: Focused on the forgetting problem, also known as catastrophic forgetting, in artificial intelligence systems. The forgetting problem refers to the tendency of neural networks and other machine learning models to forget previously learned knowledge when trained on new data.
  • Theoretical foundations and analyses of the forgetting problem.
  • Novel architectural designs and training strategies to mitigate forgetting.
  • Continual learning algorithms for overcoming catastrophic forgetting.
  • Memory consolidation and replay mechanisms for knowledge preservation.
  • Evaluation metrics and benchmarks for assessing forgetting in AI systems.
  • Applications of continual learning in domains such as computer vision, natural language processing, reinforcement learning, and robotics.
  • Neurobiological inspirations and connections to human-like learning and forgetting.
  1. Alignment and safety: As LLMs become more powerful, ensuring they behave in ways aligned with human values and intentions is critical.
  • Developing robust methods to align LLM outputs with human preferences.
  • Mitigating risks such as the generation of false or harmful information.
  • Addressing ethical concerns related to bias, fairness, and potential misuse.
  1. Hallucination in LLMs is indeed a significant research issue. Here is an overview of this problem: Hallucination in LLMs refers to the phenomenon where a model generates content that is false, inconsistent, or not supported by its training data. This is a critical issue because it can lead to the spread of misinformation and reduce the reliability of AI-generated content. Here are some key aspects of this research area:
  • Causes of hallucination: Investigating why LLMs produce hallucinations, which may be due to biases in training data, limitations in model architecture, or flaws in the training process.
  • Understanding how a model's attempt to maintain coherence in its outputs can lead to fabricated information.
  • Detection methods: Developing robust techniques to automatically identify hallucinated content in LLM outputs.
  • Creating benchmarks and evaluation metrics to measure the frequency and severity of hallucinations.
  • Mitigation strategies: Exploring methods to reduce hallucinations during training, such as improved data curation or novel training techniques.
  • Investigating post-processing techniques to filter out or correct hallucinated content.
  • Uncertainty quantification: Improving LLMs' ability to express uncertainty about their knowledge and outputs.

Prof. Dr. Hwan-Seung Yong
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • large language models
  • catastrophic forgetting
  • alignment & safety
  • hallucinations

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (2 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

18 pages, 1147 KiB  
Article
Between Truth and Hallucinations: Evaluation of the Performance of Large Language Model-Based AI Plugins in Website Quality Analysis
by Karol Król
Appl. Sci. 2025, 15(5), 2292; https://doi.org/10.3390/app15052292 - 20 Feb 2025
Viewed by 613
Abstract
Although large language models (LLMs) like the Generative Pre-trained Transformer (GPT) are growing increasingly popular, much remains to learn about their potential for website quality auditing. The article evaluates the performance of LLM AI plugins (GPT models) in website and web application auditing. [...] Read more.
Although large language models (LLMs) like the Generative Pre-trained Transformer (GPT) are growing increasingly popular, much remains to learn about their potential for website quality auditing. The article evaluates the performance of LLM AI plugins (GPT models) in website and web application auditing. The author built and tested two original ChatGPT-4o Plus (OpenAI) plugins: Website Quality Auditor (WQA) and WebGIS Quality Auditor (WgisQA). Their performance was cautiously and carefully analysed and compared to traditional auditing tools. The results demonstrated the limitations of the AI plugins, including their propensity for false outcomes. The general conclusion is that using AI tools without considering their characteristics may lead to the propagation of AI hallucinations in audit reports. The study fills in the research gap with the results on the capabilities and limitations of AI plugins in the context of auditing. It also suggests further directions for improvement. Full article
Show Figures

Figure 1

22 pages, 990 KiB  
Article
TR-GPT-CF: A Topic Refinement Method Using GPT and Coherence Filtering
by Ika Widiastuti and Hwan-Seung Yong
Appl. Sci. 2025, 15(4), 1962; https://doi.org/10.3390/app15041962 - 13 Feb 2025
Viewed by 626
Abstract
Traditional topic models are effective at uncovering patterns within large text corpora but often struggle with capturing the contextual nuances necessary for meaningful interpretation. As a result, these models may produce incoherent topics, making it challenging to achieve consistency and clarity in topic [...] Read more.
Traditional topic models are effective at uncovering patterns within large text corpora but often struggle with capturing the contextual nuances necessary for meaningful interpretation. As a result, these models may produce incoherent topics, making it challenging to achieve consistency and clarity in topic interpretation—limitations that hinder their utility for real-world applications requiring reliable insights. To overcome these challenges, we introduce a novel post-extracted topic refinement approach that uses Z-score centroid-based misaligned word detection and hybrid semantic–contextual word replacement with WordNet and GPT to replace misaligned words within topics. Evaluations across multiple datasets reveal that our approach significantly enhances topic coherence, providing a robust solution for more interpretable and semantically coherent topics. Full article
Show Figures

Figure 1

Back to TopTop