AI Academic Papers

How do scientists use the AI models in their work, and what impact will AI have on the scientific community?

Artificial Intelligence and Democracy: A Conceptual Framework

For better monitoring of AI’s impact on democracy in an interdisciplinary effort by computer and social scientists.

This paper provides an overview of AI’s historical context and evolution, emphasizing the concept of AI democratization. Current trends shaping AI democratization are analyzed, highlighting key challenges and opportunities.

Can we design artificial intelligence (AI) systems that rank our social media feeds to consider democratic values such as mitigating partisan animosity as part of their objective functions?

New research analyzes the misuse of multimodal generative AI today, in order to help build safer and more responsible technologies (2024)

The model is created to guide the prioritization of ethics in human-AI teaming by outlining the ethical teaming process, outcomes of ethical teams, and external requirements necessary to ensure ethical human-AI teams.

Description g. Our post-colonial framework unpacks AI through its (1) global, (2) meso, and (3) local layers, and further dissects how these layers are vertically integrated, each with its horizontal dependenciesoes here

The research is based on a survey of 319 knowledge workers across various professions, who provided 936 real-world examples of how they use GenAI in their work.

Building upon the findings of our integrative review, we recommend four future directions that researchers, practitioners, and policymakers can take to advance more ethical practices of AI development and use across AI value chains.

Has use cases examples on the application of the risk management framework in real-world scenarios:

This study explores the neural and behavioral consequences of LLM-assisted essay writing. Results show cognitive debt in neural, linguistic, and behavioral levels in overreliance of LLMs.

Drawing on insights from neuroscience, cognitive psychology, and learning theory, we explain how underuse of the brain’s declarative and procedural memory systems undermines reasoning, impedes learning, and diminishes productivity.

We examine recent research that asks whether current AI systems may be developing a capacity for ‘scheming’ (covertly and strategically pursuing misaligned goals).

This study underscores the dual nature of technological influence: while it fosters cultural enrichment and connectivity, it also poses risks to cultural diversity and equity.

Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity

Self-Reported Reductions in Cognitive Effort and Confidence Effects

Randomized experiments provide evidence that these types of algorithmic recommender systems change how people interact with and perceive one another in both pro-social and anti-social ways.

This paper provides a synthesis of current research trends and suggests potential directions for future research to address bias and hallucination in LLMs, considering the ongoing challenges in this field.

We demonstrate that it must be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web.

On the reliability of Large Language Models to misinformed and demographically informed prompts

How do content moderation systems across various platforms operate in indigenous and other language contexts of the Global South?

Pre-Print

Do We Really Need a Complex Agent System? Distill Embodied Agent into a Single Model

Feedback-Based Self-Learning in Large-Scale Conversational AI Agents
Published in AI Magazine

Layered MoA architecture wherein each layer comprises multiple LLM agents.

Multi-agent referee team called ChatEval to autonomously discuss and evaluate the quality of generated responses from different models

Swarm Intelligence and cyber-physical systems: Concepts, challenges and future trends

From Psychological Curiosity to Artificial Curiosity: Curiosity-Driven Learning in Artificial Intelligence Tasks

KG-RAG: Bridging the Gap Between Knowledge and Creativity

Unveiling the evolution of generative AI (GAI): a comprehensive and investigative analysis toward LLM models (2021–2024) and beyond

This architecture underscores the benefits of using generative agents as a research tool to glean new insights into real-world individual behavior.

They propose a new approach that leverages the collective strengths of multiple LLMs through a Mixture-of-Agents (MoA) methodology.

Our survey aims to serve as a valuable resource for researchers and practitioners interested in developing and deploying small yet efficient language models.

Unveiling the Potential of a Conversational Agent in Developer Support

Gorllia LLM Open source: RAFT: Adapting Language Model to Domain Specific RAG

Our study bridges the gap between information theory and architectural design, establishing entropy dynamics as a principled guide for developing efficient PI architectures.

Our framework allows researchers to discover potential mechanisms that impact life outcomes as well as the associated possibilities for personalized interventions.

Automated interpretation of electromyography using artificial intelligence to improve diagnosis in neuromuscular disorders

OPI•AID - Pioneering personalized perioperative pain management. Potentials to improve the quality of surgical procedures with targeted opioid doses for less pain and opioid-related side effects.

Critical Assessment of protein Structure Prediction (CASP14). Highly accurate protein structure prediction with AlphaFold

Advancing the Framework: Use of Health Data—A Report of a Working Conference of the American Medical Informatics Association

This paper offers approaches to policymakers, implementation scientists, clinical scientists, government regulatory bodies, and those working in the health and digital technology fields to take the lead in centering equity.

This review will discuss the relevant ethical and bias considerations in AI-ML specifically within the pathology and medical domains

Intersectional Biases in Generative Language Models and Their Psychosocial Impacts

Bias Out-of-the-Box: An Empirical Analysis of Intersectional Occupational Biases in Popular Generative Language Models

The study addresses the current state of “AI anomia”—the collective inability to name AI-related tools

They investigate the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection. Using that framework, we then perform a resume audit study to determine whether a selection of Massive Text Embedding (MTE) models are biased in resume screening scenarios.

From Model Design to Organizational Design: Complexity Redistribution and Trade-Offs in Generative AI

It highlights the importance of inclusive design and implementation of digital technologies

Artificial intelligence in education: A systematic literature review - ScienceDirect

Tutor CoPilot demonstrates how Human-AI systems can scale expertise in real-world domains, bridge gaps in skills and create a future where high-quality education is accessible to all students.

Ahead of the Curve: Implementation Challenges in Personalized Learning School Models

A recent study of 666 participants reveals a troubling connection: the more people use AI tools, the worse they perform on critical thinking tests. The impact appears strongest among young users.

OP-ED : https://the-decoder.com/frequent-ai-use-linked-to-lower-critical-thinking-scores-research-shows/

This autoethnographic study explores the need for interdisciplinary education spanning both technical and philosophical skillsn goes here

The dropout risk early detection system can be used in any online learning environment