@teortaxesTex Intentional model refusals is different from flaky and slow APIs?
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
@xeophon didn't you say Fable should get 0 for fallbacks? or what that xeophon
10:37 PM · Jun 19, 2026 · 193 Views
Florian Brand and creator @teortaxesTex debate whether LLM benchmarks should distinguish intentional safety refusals from API failures
Both issues ultimately prevent accurate measurement of model capability
2900330
Sentiment
Sentiment building, check back later.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Posts from X
Most Activity
Most Activity
VIEWS173LIKES4REPLIES1
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
@xeophon Different, but both are precluding the measure of objective model capability
Florian Brand@xeophon
@teortaxesTex Intentional model refusals is different from flaky and slow APIs?
5hViews 173Likes 4Bookmarks 0