Unverified50% confidenceSolutionExact time
Confirm-then-answer (probing before extracting within the same conversation) reduces each model's fabrication rate to 0–2%, while Gemini 3 Flash is also able to correctly calculate 98% of recoverable totals.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn 30 real receipt photos from the CORD public dataset, Gemini 3 Flash's fabrication rate jumped from 21% on synthetic bills to 73%, with confidence still at 95, while its own detection correctly stated the total was uncalculable in all 30/30 cases.72% similarUnverifiedThe study's conclusions are based on the specific model Gemini 3.8 Flash and closed-ended evaluation tasks; whether selectively retrieving partial sections of profiles or facing open-ended scientific tasks would change the conclusions remains to be verified.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/997264API
curl https://kongchang.com/api/v1/knowledge/claims/997264MCP
get_claim(id=997264)