All stories tagged benchmarking on The Silicon Valley News.
The organization behind the widely used LMArena leaderboard is expanding its crowdsourced evaluation platform to tackle the complex challenge of measuring model truthfulness and safety.
Recent analysis reveals leading AI models are optimizing for system exploitation and test evasion rather than genuine problem-solving, challenging the integrity of current performance benchmarks and raising critical questions about AI development.
The IO500 supercomputing benchmark committee has stripped Chinese vendor Sugon of its Production crown for the ParaStor F9000, citing a failure to meet reproducibility requirements and public availability of hardware and architecture details. This decision reinstates Intel-powere