Unverified50% confidenceBenchmarkExact time
Adding a rule book to the prompt drove the wild-guess rate to zero for seven of the eight models, and reduced Qwen from 59% to 7%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/11/2026
First Seen
Valid until: 10/25/2026
Sources
Related Entities
Related Claims
UnverifiedRanked by question difficulty, persistence rate was 91.4% on the 60 questions that all models answered correctly, but only 73.5% on the 52 questions that at least one model got wrong, indicating that sycophancy tracks uncertainty.72% similarUnverifiedReducing model parameters from 70B to 7B lowers both the VRAM required for inference and the per-request computation by approximately one order of magnitude.70% similarUnverifiedAdding 7 and then taking modulo 7 is equivalent to doing nothing, and can be used to shift negative remainders into the positive range.68% similarUnverifiedModels' standard accuracy on the passage-level ranking task was only 0% to 13%.68% similarUnverifiedAdding a "Do not guess" instruction to the prompt can reduce the proportion of fabricated false claims by a model from 71% to 20%.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/999561API
curl https://kongchang.com/api/v1/knowledge/claims/999561MCP
get_claim(id=999561)