Science and breakthroughs
A fix for the strawberry problem: retrofitting LLMs to read letters
By The desk · · 1 min read
research
Current state
The desk has not filed a dated note for this topic.
New benchmark: no coding agent reliably builds the agent you asked for
AgentSWE tests whether coding agents can create, edit and optimise other AI agents from a plain-language requirement. None came close, the authors report.
A topic is a tag already on a story. It is listed here when at least two stories use it.
Made with AI
Made with AIScience and breakthroughs
An interpretability researcher argues neural networks are cleaner than we think
By The desk · · 1 min read
Made with AIScience and breakthroughs
Sakana's AI reviewer catches most planted errors in papers, and far fewer real ones
By The desk · · 1 min read
Made with AIScience and breakthroughs
AI flooded math with proofs. Now it is paying for the fact-checkers.
By The desk · · 2 min read
Questions
What is on this page?
This page lists 5 stories tagged research.
What is the current state?
The desk has not filed a dated note for this topic.
Where are the sources?
Each story links the sources it used.