CAMEL-AI Agentic Research Meet Up & Live Talk

​Join us for a live talk in London & the🐫 CAMEL-AI Discord with Jonathan Cook about the paper "TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation" by Cohere.

​The paper introduces TICK (Targeted Instruct-evaluation with Checklists), a novel evaluation framework for Large Language Models (LLMs) that generates tailored checklists to assess specific instruction-following abilities. By structuring evaluations around yes/no questions, TICK improves alignment with human judgments, while STICK (Self-TICK) allows LLMs to iteratively refine their responses based on this targeted feedback, enhancing performance across diverse benchmarks.✨

​🔗 Dive into the research beforehand: https://arxiv.org/abs/2410.03608

​Event Agenda:

• ​17:00-17:30 Check-in & doors open

• ​17:30-18:00 Networking & Refreshment

• ​18:00-18:45 Live talk &  Q&A by Jonathan Cook about the paper "TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation"

• ​19:00-20:00 Live AI Demos.

• ​20:00 - Head to the pub

​You can also sign up to join us online here: https://discord.gg/vaEaBy2h