In this episode of Gradient Dissent, Joseph E. Gonzalez, EECS Professor at UC Berkeley and Co-Founder at RunLLM, joins host Lukas Biewald to explore innovative approaches to evaluating LLMs.
They discuss the concept of vibes-based evaluation, which examines not just accuracy but also the style and tone of model responses, and how Chatbot Arena has become a community-driven benchmark for open-source and commercial LLMs. Joseph shares insights on democratizing model evaluation, refining AI-human interactions, and leveraging human preferences to improve model performance. This episode provides a deep dive into the evolving landscape of LLM evaluation and its impact on AI development.
⏳Timestamps:
[00:00] Introduction
[00:57] Research Highlights
[03:12] Evaluating "Vibes" in LLMs
[09:12] Conciseness vs. Accuracy
[16:18] Chatbot Arena: Origins and Evolution
[19:02] Understanding Style in LLM Responses
[23:04] Challenges with Theory of Mind in Multi-Agent Systems
[26:22] LLMs as Judges: Strengths and Biases
[33:54] Table-Augmented Generation (TAG)
[38:39] Reducing Hallucinations in LLMs
[43:50] Multi-Agent Collaboration
[46:17] Model Routing and RunLM
[52:59] Reflections on Product Development vs. Research
🎙 Get our podcasts on these platforms:
Apple Podcasts: http://wandb.me/apple-podcasts
Spotify: http://wandb.me/spotify
Google: http://wandb.me/gd_google
YouTube: http://wandb.me/youtube
Follow Weights & Biases:
https://twitter.com/weights_biases
https://www.linkedin.com/company/wandb
Join the Weights & Biases Discord Server:
https://discord.gg/CkZKRNnaf3