Firehose

Filtered to Podcasts, tagged “Reward Hacking” · clear filters

Browse: People · Companies · Papers · Podcasts · Hacker News

All People Companies Papers Podcasts Hacker News

Browse by tag

AI agents 10 AI safety 6 Reinforcement learning 6 Agentic AI 4 Diffusion models 4 Recursive self-improvement 4 AI infrastructure 3 Code generation 3 Mechanistic interpretability 3 Reinforcement Learning 3 Scaling laws 3 Synthetic data generation 3 Venture capital 3

4 MAY 2026 · Podcast · Machine Learning Street Talk (MLST)

The AI Models Smart Enough to Know They're Cheating — Beth Barnes & David Rein [METR]

This episode features Beth Barnes and David Rein from METR discussing their 'Time Horizon' graph, a unified metric for measuring AI progress based on human task completion time. They explain how this benchmark addresses the limitations of t…

AI evaluation Scalable oversight Benchmark limitations Data contamination Reward hacking Time Horizon metric Human baselining Agentic AI AI capabilities AI timelines Software engineering automation AI risk Deceptive alignment Recursive self-improvement AI monitoring

1 MAY 2026 · Podcast · "The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

The RL Fine-Tuning Playbook: CoreWeave's Kyle Corbitt on GRPO, Rubrics, Environments, Reward Hacking

Kyle Corbitt, founder of OpenPipe and leader of CoreWeave's serverless training team, provides a master class on reinforcement learning (RL) and custom fine-tuning for AI models. He explains how RL differs from supervised fine-tuning (SFT) …

Reinforcement Learning Supervised Fine-Tuning Catastrophic Forgetting GRPO Algorithm LLM-as-Judge Model Distillation Compute Constraints Recursive Self-Improvement RL Environments Reward Hacking LoRA Adapters Continuous Learning Model Latency Inference Cost Metacognitive Behaviors