Paper Review Reading List

The rise of AI has led to a rapid increase in published research, and this trend is only accelerating. In an attempt to stay informed, I’ll be posting brief paper summaries with key takeaways and any questions or topics I want to dig into further. All posts will be listed under the Paper Reviews category. I’ll also keep adding to the backlog here as I find more interesting stuff. I’m sure this page will become quite the behemoth in no time.

Backlog:

Safety & Alignment

• Evaluating Frontier Models for Dangerous Capabilities
• Cultural Evolution of Cooperation among LLM Agents
• Frontier AI systems have surpassed the self-replicating red line
• Frontier Models are Capable of In-context Scheming
• Frontier AI Regulation: Managing Emerging Risks to Public Safety
• ALIGNMENT FAKING IN LARGE LANGUAGE MODELS
• AI safety via debate

Agentic Systems

• Agents for self-driving laboratories applied to quantum computing
• ChatDev: Communicative Agents for Software Development

Audio & Voice

• Sketch2Sound: Controllable Audio Generation via Time-Varying Signals and Sonic Imitations

Prompting

• A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT
• The Prompt Report: A Systematic Survey of Prompting Techniques
• A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications

Misc

• Human confidence in artificial intelligence and in themselves: The evolution and impact of confidence on adoption of AI advice