This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
Tamas Nagy
tamasnagy.com
did:plc:ugakxorq4mwoxgyj7scfp4ef
Interesting work that shows that frontier models are basically not improving in reliability despite them appearing to do better on benchmarks. Reliability is key to trust. Hard to imagine LLMs being trusted with critical operations until this is solved, *if* it can be.
arxiv.org/pdf/2602.16666
2026-07-21T22:59:46.275Z