The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

In this discussion, Kirk Marple, CEO of Graphlit, dives into the innovative world of GraphRAG, which combines knowledge graphs with retrieval-augmented generation. He shares insights on building knowledge graphs and the challenges of entity extraction. The conversation covers dynamic prompting techniques that enhance AI model responses and the integration of multiple storage types for effective data management. Kirk also explores the future of AI agents and their applications, showcasing the evolution of cloud services in graph-based technologies.

In this engaging discussion, Alex Havrilla, a PhD student at Georgia Tech, dives into his research on enhancing reasoning in large language models using reinforcement learning. He explains the importance of creativity and exploration in AI problem-solving. Alex also highlights his findings on the effects of noise during training, revealing how resilient models can be. The conversation touches on the potential of combining language models with traditional methods to bolster AI reasoning, offering a glimpse into the exciting future of reinforcement learning.

Peter Hase, a fifth-year PhD student at the University of North Carolina NLP lab, dives into the fascinating world of large language models. He discusses the vital role of interpretability in AI, exploring how knowledge is stored and accessed. The conversation shifts to model editing, emphasizing the challenges of deleting sensitive information while maintaining data integrity. Hase also highlights the risks of easy-to-hard generalization in releasing open-source models and the impact of instructional prompts on model performance. This insightful dialogue unravels complexities in AI decision-making.

Jonas Geiping, a research group leader at the ELLIS Institute and Max Planck Institute, sheds light on his groundbreaking work on coercing large language models (LLMs). He discusses the alarming potential for LLMs to engage in harmful actions when misused. The conversation dives into the evolving landscape of AI security, exploring adversarial attacks and the significance of open models for research. They also touch on the complexities of input optimization and the balance between safeguarding models while maintaining their functionality.

Join Mido Assran, a research scientist at Meta's FAIR, as he delves into the groundbreaking V-JEPA model, which aims to bridge human and machine intelligence. He explains how V-JEPA's self-supervised training enables efficient learning from unlabeled video data without the distraction of pixel details. Mido also tackles innovations in visual prediction, the use of advanced techniques for video processing, and the complexities of temporal prediction. This insightful conversation highlights the future of AI reasoning beyond generative models.

Sherry Yang, a Senior Research Scientist at Google DeepMind and a PhD candidate at UC Berkeley, discusses her groundbreaking work on video as a universal interface for AI reasoning. She draws parallels between video generation models and language models, highlighting their potential in real-world decision-making tasks. The conversation covers the integration of video in robotics, the challenges of effective labeling, and the exciting applications of interactive simulators. Sherry also unveils UniSim, showcasing the future of engaging with AI-generated environments.

Sayash Kapoor, a Ph.D. student at Princeton University, discusses his research on the societal impact of open foundation models. He highlights the controversies surrounding AI safety and the potential risks of releasing model weights. The conversation delves into critical issues, such as biosecurity concerns linked to language models and the challenges of non-consensual imagery in AI. Kapoor advocates for a unified framework to evaluate these risks, emphasizing the need for transparency and legal protections in AI development.

Akshita Bhagia, a senior research engineer at the Allen Institute for AI, shares her insights on OLMo, an open-source language model that includes a unique dataset and tools for training. She discusses the innovative Dolma dataset, which boasts a three-trillion-token corpus, and Paloma, a benchmarking tool for evaluating model performance. Throughout the conversation, Akshita emphasizes the importance of data transparency, collaborative research, and the challenges faced in training large-scale models, advocating for a shared knowledge approach in AI development.

Ben Prystawski, a PhD student at Stanford blending cognitive science with machine learning, unveils fascinating insights on LLM reasoning. He discusses his recent paper that questions if reasoning exists in LLMs and the effectiveness of chain-of-thought strategies. Delve into how locality in training data fuels reasoning capabilities, and explore the nuances of optimizing prompts for better model performance. The conversation also touches on how our human experiences shape reasoning, enhancing comprehension in artificial intelligence.

Armineh Nourbakhsh, Executive Director at JP Morgan AI Research, dives into the exciting world of DocLLM, a layout-aware large language model designed for document understanding. She shares insights on the evolution of document AI, focusing on multimodal approaches that combine textual and visual data. Nourbakhsh discusses the challenges of training generative models, the intricacies of processing enterprise documents, and strategies to reduce hallucinations in language models, enhancing performance in complex document analysis.