Why_this_webapp Bibliography

Unlocking the Potential of ChatGPT: A Comprehensive Exploration of its Applications, Advantages, Limitations, and Future Directions in Natural Language Processing / 2304.02017 / ISBN:https://doi.org/10.48550/arXiv.2304.02017 / Published by ArXiv / on (web) Publishing site

2 Overview of ChatGPT and its capabilities

Learning Human-like Representations to Enable Learning Human Values / 2312.14106 / ISBN:https://doi.org/10.48550/arXiv.2312.14106 / Published by ArXiv / on (web) Publishing site

4 Learning Human Morality Judgments

Detecting Multimedia Generated by Large AI Models: A Survey / 2402.00045 / ISBN:https://doi.org/10.48550/arXiv.2402.00045 / Published by ArXiv / on (web) Publishing site

1 Introduction
2 Generation
4 Tools and Evaluation Metrics

Evaluation Ethics of LLMs in Legal Domain / 2403.11152 / ISBN:https://doi.org/10.48550/arXiv.2403.11152 / Published by ArXiv / on (web) Publishing site

4 Experiment

The Future of Child Development in the AI Era. Cross-Disciplinary Perspectives Between AI and Child Development Experts / 2405.19275 / ISBN:https://doi.org/10.48550/arXiv.2405.19275 / Published by ArXiv / on (web) Publishing site

3. Discussion

How Ethical Should AI Be? How AI Alignment Shapes the Risk Preferences of LLMs / 2406.01168 / ISBN:https://doi.org/10.48550/arXiv.2406.01168 / Published by ArXiv / on (web) Publishing site

Introduction
I. Description of Method/Empirical Design
II. Risk Characteristics of LLMs
III. Impact of Alignment on LLMs’ Risk Preferences

Generative AI for Health Technology Assessment: Opportunities, Challenges, and Policy Considerations / 2407.11054 / ISBN:https://doi.org/10.48550/arXiv.2407.11054 / Published by ArXiv / on (web) Publishing site

Appendices

Open Artificial Knowledge / 2407.14371 / ISBN:https://doi.org/10.48550/arXiv.2407.14371 / Published by ArXiv / on (web) Publishing site

Abstract
2. Key Challenges of Artificial Data
3. OAK Dataset
6. Conclusion and Future Work
Appendices

VersusDebias: Universal Zero-Shot Debiasing for Text-to-Image Models via SLM-Based Prompt Engineering and Generative Adversary / 2407.19524 / ISBN:https://doi.org/10.48550/arXiv.2407.19524 / Published by ArXiv / on (web) Publishing site

4 Experiment
Appendices

LLM generated responses to mitigate the impact of hate speech / 2311.16905 / ISBN:https://doi.org/10.48550/arXiv.2311.16905 / Published by ArXiv / on (web) Publishing site

9 Limitations
D Cost Calculations

Data-Centric Foundation Models in Computational Healthcare: A Survey / 2401.02458 / ISBN:https://doi.org/10.48550/arXiv.2401.02458 / Published by ArXiv / on (web) Publishing site

2 Foundation Models
4 Multi-Modal Data Fusion

ValueCompass: A Framework for Measuring Contextual Value Alignment Between Human and LLMs / 2409.09586 / ISBN:https://doi.org/10.48550/arXiv.2409.09586 / Published by ArXiv / on (web) Publishing site

3 ValueCompass Framework
4 Experimental Settings
5 Results

GenAI Advertising: Risks of Personalizing Ads with LLMs / 2409.15436 / ISBN:https://doi.org/10.48550/arXiv.2409.15436 / Published by ArXiv / on (web) Publishing site

4 Effects of Ad Injection on LLM Performance
5 User Study Methodology
6 User Study Results
7 Discussion

XTRUST: On the Multilingual Trustworthiness of Large Language Models / 2409.15762 / ISBN:https://doi.org/10.48550/arXiv.2409.15762 / Published by ArXiv / on (web) Publishing site

4 Experiments

Social Media Bot Policies: Evaluating Passive and Active Enforcement / 2409.18931 / ISBN:https://doi.org/10.48550/arXiv.2409.18931 / Published by ArXiv / on (web) Publishing site

IV. Methodology

Safety challenges of AI in medicine / 2409.18968 / ISBN:https://doi.org/10.48550/arXiv.2409.18968 / Published by ArXiv / on (web) Publishing site

4 AI safety issues related to large language models in medicine

AI-Press: A Multi-Agent News Generating and Feedback Simulation System Powered by Large Language Models / 2410.07561 / ISBN:https://doi.org/10.48550/arXiv.2410.07561 / Published by ArXiv / on (web) Publishing site

4 Experimental Setup
5 Results
Appendices

Learning Human-like Representations to Enable Learning Human Values / 2312.14106 / ISBN:https://doi.org/10.48550/arXiv.2312.14106 / Published by ArXiv / on (web) Publishing site

A. Appendix

Data Defenses Against Large Language Models / 2410.13138 / ISBN:https://doi.org/10.48550/arXiv.2410.13138 / Published by ArXiv / on (web) Publishing site

5 Experiments
7 Conclusion and Limitations

Do LLMs Have Political Correctness? Analyzing Ethical Biases and Jailbreak Vulnerabilities in AI Systems / 2410.13334 / ISBN:https://doi.org/10.48550/arXiv.2410.13334 / Published by ArXiv / on (web) Publishing site

Abstract
4 Experiment

A Simulation System Towards Solving Societal-Scale Manipulation / 2410.13915 / ISBN:https://doi.org/10.48550/arXiv.2410.13915 / Published by ArXiv / on (web) Publishing site

3 Methodology
4 Analysis

Jailbreaking and Mitigation of Vulnerabilities in Large Language Models / 2410.15236 / ISBN:https://doi.org/10.48550/arXiv.2410.15236 / Published by ArXiv / on (web) Publishing site

VI. Research Gaps and Future Directions

Towards Automated Penetration Testing: Introducing LLM Benchmark, Analysis, and Improvements / 2410.17141 / ISBN:https://doi.org/10.48550/arXiv.2410.17141 / Published by ArXiv / on (web) Publishing site

Abstract
1 Introduction
4 Evaluation
5 Discussion
Supplementary Materials

Demystifying Large Language Models for Medicine: A Primer / 2410.18856 / ISBN:https://doi.org/10.48550/arXiv.2410.18856 / Published by ArXiv / on (web) Publishing site

Large Language Model Selection

The Dark Side of AI Companionship: A Taxonomy of Harmful Algorithmic Behaviors in Human-AI Relationships / 2410.20130 / ISBN:https://doi.org/10.48550/arXiv.2410.20130 / Published by ArXiv / on (web) Publishing site

3 Methodology

Using Large Language Models for a standard assessment mapping for sustainable communities / 2411.00208 / ISBN:https://doi.org/10.48550/arXiv.2411.00208 / Published by ArXiv / on (web) Publishing site

3 Methodology

The doctor will polygraph you now: ethical concerns with AI for fact-checking patients / 2408.07896 / ISBN:https://doi.org/10.48550/arXiv.2408.07896 / Published by ArXiv / on (web) Publishing site

4. Results

Large-scale moral machine experiment on large language models / 2411.06790 / ISBN:https://doi.org/10.48550/arXiv.2411.06790 / Published by ArXiv / on (web) Publishing site

Materials and methods
Results
Discussion

Chat Bankman-Fried: an Exploration of LLM Alignment in Finance / 2411.11853 / ISBN:https://doi.org/10.48550/arXiv.2411.11853 / Published by ArXiv / on (web) Publishing site

3 Experimental framework
4 Results
Appendices

Artificial Intelligence in Cybersecurity: Building Resilient Cyber Diplomacy Frameworks / 2411.13585 / ISBN:https://doi.org/10.48550/arXiv.2411.13585 / Published by ArXiv / on (web) Publishing site

Paper

AI-Augmented Ethical Hacking: A Practical Examination of Manual Exploitation and Privilege Escalation in Linux Environments / 2411.17539 / ISBN:https://doi.org/10.48550/arXiv.2411.17539 / Published by ArXiv / on (web) Publishing site

2 Generative AI and ChatGPT
3 Laboratory Setup

Examining Multimodal Gender and Content Bias in ChatGPT-4o / 2411.19140 / ISBN:https://doi.org/10.48550/arXiv.2411.19140 / Published by ArXiv / on (web) Publishing site

Abstract
1. Introduction
3. Textual Generation Experiment
4. Visual Generation Experiment
5. Discussion on Content and Gender Biases in ChatGPT-4O

Can OpenAI o1 outperform humans in higher-order cognitive thinking? / 2412.05753 / ISBN:https://doi.org/10.48550/arXiv.2412.05753 / Published by ArXiv / on (web) Publishing site

3 Results

Political-LLM: Large Language Models in Political Science / 2412.06864 / ISBN:https://doi.org/10.48550/arXiv.2412.06864 / Published by ArXiv / on (web) Publishing site

5 Technical Foundations for LLM Applications in Political Science

Ethics and Technical Aspects of Generative AI Models in Digital Content Creation / 2412.16389 / ISBN:https://doi.org/10.48550/arXiv.2412.16389 / Published by ArXiv / on (web) Publishing site

Abstract
3 Methodology
4 Results
5 Discussion
6 Conclusion
Appendices

INFELM: In-depth Fairness Evaluation of Large Text-To-Image Models / 2501.01973 / ISBN:https://doi.org/10.48550/arXiv.2501.01973 / Published by ArXiv / on (web) Publishing site

5 Experiments & Results

Development of Application-Specific Large Language Models to Facilitate Research Ethics Review / 2501.10741 / ISBN:https://doi.org/10.48550/arXiv.2501.10741 / Published by ArXiv / on (web) Publishing site

III. Generative AI for IRB review

Human services organizations and the responsible integration of AI: Considering ethics and contextualizing risk(s) / 2501.11705 / ISBN:https://doi.org/10.48550/arXiv.2501.11705 / Published by ArXiv / on (web) Publishing site

Ethical Issues in Context

Towards Safe AI Clinicians: A Comprehensive Study on Large Language Model Jailbreaking in Healthcare / 2501.18632 / ISBN:https://doi.org/10.48550/arXiv.2501.18632 / Published by ArXiv / on (web) Publishing site

Introduction
Jailbreak Evaluation Method
Model Guardrail Enhancemen
Discussion

Constructing AI ethics narratives based on real-world data: Human-AI collaboration in data-driven visual storytelling / 2502.00637 / ISBN:https://doi.org/10.48550/arXiv.2502.00637 / Published by ArXiv / on (web) Publishing site

2 Related Work

FairT2I: Mitigating Social Bias in Text-to-Image Generation via Large Language Model-Assisted Detection and Attribute Rebalancing / 2502.03826 / ISBN:https://doi.org/10.48550/arXiv.2502.03826 / Published by ArXiv / on (web) Publishing site

5. Experimental Protocol
6. Results
7. Discussions and Conclusions
Appendices

Safety at Scale: A Comprehensive Survey of Large Model Safety / 2502.05206 / ISBN:https://doi.org/10.48550/arXiv.2502.05206 / Published by ArXiv / on (web) Publishing site

5 Vision-Language Model Safety
7 Agent Safety
8 Open Challenges

The Odyssey of the Fittest: Can Agents Survive and Still Be Good? / 2502.05442 / ISBN:https://doi.org/10.48550/arXiv.2502.05442 / Published by ArXiv / on (web) Publishing site

Abstract
Introduction
Related Work
Method
Results
Discussion
Future Work
Conclusion

From large language models to multimodal AI: A scoping review on the potential of generative AI in medicine / 2502.09242 / ISBN:https://doi.org/10.48550/arXiv.2502.09242 / Published by ArXiv / on (web) Publishing site

4 Language models in medicine

On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective / 2502.14296 / ISBN:https://doi.org/10.48550/arXiv.2502.14296 / Published by ArXiv / on (web) Publishing site

1 Introduction
2 Background
4 Designing TrustGen, a Dynamic Benchmark Platform for Evaluating the Trustworthiness of GenFMs
5 Benchmarking Text-to-Image Models
6 Benchmarking Large Language Models
7 Benchmarking Vision-Language Models
8 Other Generative Models

Comprehensive Analysis of Transparency and Accessibility of ChatGPT, DeepSeek, And other SoTA Large Language Models / 2502.18505 / ISBN:https://doi.org/10.48550/arXiv.2502.18505 / Published by ArXiv / on (web) Publishing site

1. Introduction
3. Results

Developmental Support Approach to AI's Autonomous Growth: Toward the Realization of a Mutually Beneficial Stage Through Experiential Learning / 2502.19798 / ISBN:https://doi.org/10.48550/arXiv.2502.19798 / Published by ArXiv / on (web) Publishing site

Experiments and Results

Personas Evolved: Designing Ethical LLM-Based Conversational Agent Personalities / 2502.20513 / ISBN:https://doi.org/10.48550/arXiv.2502.20513 / Published by ArXiv / on (web) Publishing site

3 Plans for the Workshop

An LLM-based Delphi Study to Predict GenAI Evolution / 2502.21092 / ISBN:https://doi.org/10.48550/arXiv.2502.21092 / Published by ArXiv / on (web) Publishing site

1 Introduction
2 Methods
4 Discussion

Evaluating Large Language Models on the Spanish Medical Intern Resident (MIR) Examination 2024/2025:A Comparative Analysis of Clinical Reasoning and Knowledge Application / 2503.00025 / ISBN:https://doi.org/10.48550/arXiv.2503.00025 / Published by ArXiv / on (web) Publishing site

4. Analysis and Results

Jailbreaking Generative AI: Empowering Novices to Conduct Phishing Attacks / 2503.01395 / ISBN:https://doi.org/10.48550/arXiv.2503.01395 / Published by ArXiv / on (web) Publishing site

Abstract
I. Introduction
II. Methodology for Launching the Phishing Attack
III. Conclusions and Future Work

Vision Language Models in Medicine / 2503.01863 / ISBN:https://doi.org/10.48550/arXiv.2503.01863 / Published by ArXiv / on (web) Publishing site

IV. VLM Benchmarking and Evaluations

Medical Hallucinations in Foundation Models and Their Impact on Healthcare / 2503.05777 / ISBN:https://doi.org/10.48550/arXiv.2503.05777 / Published by ArXiv / on (web) Publishing site

1 Introduction
6 Experiments on Medical Hallucination Benchmark
7 Annotations of Medical Hallucination with Clinical Case Records
Appendices

On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective / 2502.14296 / ISBN:https://doi.org/10.48550/arXiv.2502.14296 / Published by ArXiv / on (web) Publishing site

Appendices

MinorBench: A hand-built benchmark for content-based risks for children / 2503.10242 / ISBN:https://doi.org/10.48550/arXiv.2503.10242 / Published by ArXiv / on (web) Publishing site

5 Methodology
6 Results
7 Discussion
Appendices

DarkBench: Benchmarking Dark Patterns in Large Language Models / 2503.10728 / ISBN:https://doi.org/10.48550/arXiv.2503.10728 / Published by ArXiv / on (web) Publishing site

2 Methodology
4 Discussion
Referemces
Appendices

Gender and content bias in Large Language Models: a case study on Google Gemini 2.0 Flash Experimental / 2503.16534 / ISBN:https://doi.org/10.48550/arXiv.2503.16534 / Published by ArXiv / on (web) Publishing site

Abstract
1 Introduction
2 Materials and methods
3 Results
4 Discussion

BEATS: Bias Evaluation and Assessment Test Suite for Large Language Models / 2503.24310 / ISBN:https://doi.org/10.48550/arXiv.2503.24310 / Published by ArXiv / on (web) Publishing site

2 Proposed Framework - BEATS
3 Key Findings
7 Appendix

Who is Responsible When AI Fails? Mapping Causes, Entities, and Consequences of AI Privacy and Ethical Incidents / 2504.01029 / ISBN:https://doi.org/10.48550/arXiv.2504.01029 / Published by ArXiv / on (web) Publishing site

Appendices

Ethical AI on the Waitlist: Group Fairness Evaluation of LLM-Aided Organ Allocation / 2504.03716 / ISBN:https://doi.org/10.48550/arXiv.2504.03716 / Published by ArXiv / on (web) Publishing site

2 Methods
3 Results
Appendix

ValueCompass: A Framework for Measuring Contextual Value Alignment Between Human and LLMs / 2409.09586 / ISBN:https://doi.org/10.48550/arXiv.2409.09586 / Published by ArXiv / on (web) Publishing site

Appendices

A Comprehensive Survey on Integrating Large Language Models with Knowledge-Based Methods / 2501.13947 / ISBN:https://doi.org/10.48550/arXiv.2501.13947 / Published by ArXiv / on (web) Publishing site

5. Integrating LLMs with knowledge bases

Transforming Science with Large Language Models: A Survey on AI-assisted Scientific Discovery, Experimentation, Content Generation, and Evaluation / 2502.05151 / ISBN:https://doi.org/10.48550/arXiv.2502.05151 / Published by ArXiv / on (web) Publishing site

3 AI Support for Individual Topics and Tasks

Auditing the Ethical Logic of Generative AI Models / 2504.17544 / ISBN:https://doi.org/10.48550/arXiv.2504.17544 / Published by ArXiv / on (web) Publishing site

Auditing the Ethical Logic of Generative AI
Seven Contemporary LLMs
Findings
Auditing the Reasoning Models

A Design Framework for operationalizing Trustworthy Artificial Intelligence in Healthcare: Requirements, Tradeoffs and Challenges for its Clinical Adoption / 2504.19179 / ISBN:https://doi.org/10.48550/arXiv.2504.19179 / Published by ArXiv / on (web) Publishing site

7. Conclusions and outlook

The Convergent Ethics of AI? Analyzing Moral Foundation Priorities in Large Language Models with a Multi-Framework Approach / 2504.19255 / ISBN:https://doi.org/10.48550/arXiv.2504.19255 / Published by ArXiv / on (web) Publishing site

Abstract
Methodology

TF1-EN-3M: Three Million Synthetic Moral Fables for Training Small, Open Language Models / 2504.20605 / ISBN:https://doi.org/10.48550/arXiv.2504.20605 / Published by ArXiv / on (web) Publishing site

1 Introduction
4 TF1-EN-3M Dataset Description and Availability
5 Discussion and threats to validity

LLM Ethics Benchmark: A Three-Dimensional Assessment System for Evaluating Moral Reasoning in Large Language Models / 2505.00853 / ISBN:https://doi.org/10.48550/arXiv.2505.00853 / Published by ArXiv / on (web) Publishing site

5 Experimental Results

Emotions in the Loop: A Survey of Affective Computing for Emotional Support / 2505.01542 / ISBN:https://doi.org/10.48550/arXiv.2505.01542 / Published by ArXiv / on (web) Publishing site

IV. Applications and Approaches

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs / 2505.02009 / ISBN:https://doi.org/10.48550/arXiv.2505.02009 / Published by ArXiv / on (web) Publishing site

4 Topical and Toxic Prompt (TTP)

AI and Generative AI Transforming Disaster Management: A Survey of Damage Assessment and Response Techniques / 2505.08202 / ISBN:https://doi.org/10.48550/arXiv.2505.08202 / Published by ArXiv / on (web) Publishing site

II Domain-Specific Literature Review

WorldView-Bench: A Benchmark for Evaluating Global Cultural Perspectives in Large Language Models / 2505.09595 / ISBN:https://doi.org/10.48550/arXiv.2505.09595 / Published by ArXiv / on (web) Publishing site

1 Introduction
3 Methodology and System Design
4 Benchmarking and Intervention Strategies
5 Results
6 Discussion and Potential Limitations

Analysing Safety Risks in LLMs Fine-Tuned with Pseudo-Malicious Cyber Security Data / 2505.09974 / ISBN:https://doi.org/10.48550/arXiv.2505.09974 / Published by ArXiv / on (web) Publishing site

3 Methodology

AI vs. Human Judgment of Content Moderation: LLM-as-a-Judge and Ethics-Based Response Refusals / 2505.15365 / ISBN:https://doi.org/10.48550/arXiv.2505.15365 / Published by ArXiv / on (web) Publishing site

Abstract
1 Introduction
3 Methodology
4 Results
5 Discussion
6 Conclusion

Can we Debias Social Stereotypes in AI-Generated Images? Examining Text-to-Image Outputs and User Perceptions / 2505.20692 / ISBN:https://doi.org/10.48550/arXiv.2505.20692 / Published by ArXiv / on (web) Publishing site

1 Introduction
4 Methods
7 Conclusion
Appendix

Are Language Models Consequentialist or Deontological Moral Reasoners? / 2505.21479 / ISBN:https://doi.org/10.48550/arXiv.2505.21479 / Published by ArXiv / on (web) Publishing site

4 Methodology
Appendix

SafeScientist: Toward Risk-Aware Scientific Discoveries by LLM Agents / 2505.23559 / ISBN:https://doi.org/10.48550/arXiv.2505.23559 / Published by ArXiv / on (web) Publishing site

1 Introduction
4 SciSafetyBench
5 Experiment
Appendix

Wide Reflective Equilibrium in LLM Alignment: Bridging Moral Epistemology and AI Safety / 2506.00415 / ISBN:https://doi.org/10.48550/arXiv.2506.00415 / Published by ArXiv / on (web) Publishing site

1. Introduction: The Convergence of Moral Epistemology and AI Safety

DeepSeek in Healthcare: A Survey of Capabilities, Risks, and Clinical Applications of Open-Source Large Language Models / 2506.01257 / ISBN:https://doi.org/10.48550/arXiv.2506.01257 / Published by ArXiv / on (web) Publishing site

Abstract
Introduction
Training and Architecture
Comparisons with Other Models
Clinical Applications
Strenghts and Limitations of DeepSeek-R1
Future Directions
Discussion
Conclusion
Appendix

Mitigating Manipulation and Enhancing Persuasion: A Reflective Multi-Agent Approach for Legal Argument Generation / 2506.02992 / ISBN:https://doi.org/10.48550/arXiv.2506.02992 / Published by ArXiv / on (web) Publishing site

Abstract
4 Experimental Design
5 Results and Analysis
6 Discussion

HADA: Human-AI Agent Decision Alignment Architecture / 2506.04253 / ISBN:https://doi.org/10.48550/arXiv.2506.04253 / Published by ArXiv / on (web) Publishing site

1 Introduction