Steven AI Talk

https://arxiv.org/html/2504.20879v1The Leaderboard Illusion Measuring progress is crucial for any scientific field, and in Artificial Intelligence, benchmarks play a vital role. Recently, Chatbot Arena has become the most influential leaderboard for ranking the capabilities of large language models (LLMs), significantly impacting media perception, industry trends, and academic research. However, a...

AI PodcastFin-R1: Financial Reasoning with a Lightweight Language Model https://huggingface.co/SUFE-AIFLM-Lab/Fin-R1

AI generated podcastMechanistic Interpretability via Cross-Layer Feature Attribution Graphsref: https://transformer-circuits.pub/2025/attribution-graphs/methods.html

AI generated podcastTracing the Thoughts of a Large Language ModelRef: https://www.anthropic.com/news/tracing-thoughts-language-model

Source:https://huggingface.co/microsoft/Magma-8Bhttps://github.com/microsoft/MagmaDigital and Physical Worlds: Magma is the first-ever foundation model for multimodal AI agents, designed to handle complex interactions across both virtual and real environments! Versatile Capabilities: Magma as a single model not only possesses generic image and videos understanding ability, but also generate goal-d...

Experimental Economics_ Core Concepts and Applications

The Psychology and Economics of Incentives

Market Experiments, Asymmetric Information, and Auction Theory

Behavioral Economics_ Trust, Motivation, Willpower, and Gender

Experimental Economics_ Game Theory, Trust, and Decision Making

Game Theory_ Strategic Decisions, Equilibria, and Experimental Data

Beyond Homo Economicus_ Social Preferences and Game Theory

Risk, Time Preferences, and Experimental Methodology

Experimental Economics_ Design, Validity, and Types

Behavioral Economics_ Heuristics, Prospect Theory, and the Endowment Effect

Experimental and Behavioral Economics_ An Introduction

AI generated podcast Ref: NASA-Examining Proposed Uses of LLMs to Produce or Assess Assurance Arguments

hallucination and rag - debate version podcast

hallucination and rag - executive version podcast

hallucination and rag - summary version podcast

AI generated podcast: International AI Safety Report

AI generated podcast: Transfusion A Multi Modal Generative Model

AI generated conversation discussing EU compass, new initiative to boost competitiveness and innovation of EU.

### AI Safety - "AirBench 2024: The Future of Responsible AI 🚦🤖" **"How Safe is AI? Deep Dive into AirBench 2024 🚦🤖"** --- #### **Video Description**: 🌍 **Welcome to the future of AI safety!** In this episode, we take an in-depth look at **AirBench 2024**, a groundbreaking benchmark that evaluates AI models based on real-world **risk categories**, laws, and corporate policies. Discover: - How...

### AI generated podcast: 🎥 Unlocking the Power of Kolmogorov-Arnold Networks (CANs) 🌟 | Neural Networks Deep Dive ### Video Description: 🌟 **Welcome to our latest deep dive!** 🌟 In this episode, we explore the fascinating world of **Kolmogorov-Arnold Networks (CANs)**—a groundbreaking neural network architecture. From their unique use of splines to their efficiency, interpretability, and spec...