Unverified50% confidenceBenchmarkExact time
On competition problems, VERS loses 3-7 percentage points of accuracy across all model sizes.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/3/2026
First Seen
Valid until: 2/1/2027
Sources
Related Entities
Related Claims
UnverifiedVERS reduces the number of generated tokens by 45.8% on the full GSM8K test set, with accuracy dropping by only 0.82 percentage points.75% similarUnverifiedWhen using the multi-agent verification process, MARCH's accuracy was suppressed to 4.4%, whereas when the same model was directly asked which solution was better, accuracy reached 43.7%.70% similarVerifiedINT8 quantization loses approximately 0.5% accuracy while halving VRAM usage; INT4 quantization compresses VRAM to 1/4 of the original with an accuracy loss of approximately 1–3%.69% similarUnverifiedFine-tuned DistilBERT achieves F1=0.9993 on in-distribution testing, but loses 12.9 mean average precision points under topic shift, a greater degradation than the linear model's 5.2 points.69% similarUnverifiedOn TriviaQA, the error rate of answered questions (under the exact match scoring criterion) decreased from 23.0% to 14.5%.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1130565API
curl https://kongchang.com/api/v1/knowledge/claims/1130565MCP
get_claim(id=1130565)