AI Research Radar
Friday, May 22: no frontier papers out of 112 read.
Frontier
Papers rated as moving the field
How we chooseNone of the 6 cleared papers that day was rated frontier.
This week in research
May 16–22: 56 cleared, 2 frontier
The week’s papers have not been sorted into fields yet.
Models under test
Papers this week that name the model
Labs publishing
Papers this week with an author at the lab
Also cleared
6 papers
- LLM Agents Already Know When to Call Tools -- Even Without ReasoningCode
- HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine
- CR4T: Rewrite-Based Guardrails for Adolescent LLM Safety
- Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems
- Boundary-targeted Membership Inference Attacks on Safety Classifiers
- Evaluating Commercial AI Chatbots as News Intermediaries