Advances in Human-Centric AI and Language Models

The field of artificial intelligence and language models is moving towards more human-centric and interactive approaches. Researchers are exploring ways to make AI systems more transparent, explainable, and aligned with human goals and values. One notable direction is the development of systems that can infer and adapt to user intentions, providing more effective and personalized support. Another area of focus is the creation of frameworks and tools that enable more efficient and effective fine-tuning of large language models, allowing them to better capture subtle nuances and complexities of human language. Noteworthy papers in this area include: PromptFlow, which proposes a modular training framework for prompt engineering, and Just-In-Time Objectives, which demonstrates an architecture for automatically inducing user objectives and steering downstream AI systems. These advancements have the potential to significantly improve the usability and effectiveness of AI systems in a wide range of applications.

Sources

Informative Keyboard and its Application to Raise Awareness of Smartphone Use

PromptFlow: Training Prompts Like Neural Networks

Data-Model Co-Evolution: Growing Test Sets to Refine LLM Behavior

Personalized Learning Path Planning with Goal-Driven Learner State Modeling

Generative AI in Heritage Practice: Improving the Accessibility of Heritage Guidance

PAGE: Prompt Augmentation for text Generation Enhancement

Technological Devices and Their Negative Effects on Health

State Your Intention to Steer Your Attention: An AI Assistant for Intentional Digital Living

Just-In-Time Objectives: A General Approach for Specialized AI Interactions

Boosting Instruction Following at Scale

Built with on top of