The Lock-in Hypothesis: Stagnation by Algorithm
arxiv.org · 6,045 words · saved by 1 readers
N/A
The Lock-in Hypothesis: Stagnation by Algorithm Tianyi Alex Qiu * 1 Zhonghao He * 2 Tejasveer Chugh 3 Max Kleiman-Weiner 4 Abstract 2024; Peterson, 2024; Jakesch et al., 2023). However there The training and deployment of large language remains a key gap in prior analyses. First, existing studies models (LLMs)…
related reading
- The Persona Selection Model: Why AI Assistants might Behave like Humansalignment.anthropic.com
- Guardian Angels: LLM Personalization for Productivity and Security · Gwern.netgwern.net
- The Waluigi Effect (mega-post) — LessWronglesswrong.com
- LLM Daydreaming · Gwern.netgwern.net
- As Rocks May Think | Eric Jangevjang.com
- llm assistant personas seem increasingly incoherent (some subjective observations) — LessWronglesswrong.com
- RLHF: Reinforcement Learning from Human Feedbackhuyenchip.com
- The Dark Forest and Generative AImaggieappleton.com
- Modifying LLM Beliefs with Synthetic Document Finetuningalignment.anthropic.com
- Simulated Users & Sad LLMs1a3orn.com
- LLMs as Cognitive Viruses: Population Dynamicsemergentmind.com
- Feedback Loops With Language Models Drive In-Context Reward Hackingarxiv.org