Unverified50% confidenceBenchmarkExact time
4-bit quantization changed context-free multiple-choice accuracy within a range of -3.2 to +1.6 percentage points.
1
Sources
50%
Confidence
Long-term
Relevance
11/6/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAfter re-evaluating with non-overlapping anchors, the net advantage instead grew from +4.1pp to +5.5pp, but significance dropped from 2.4 standard errors to 1.4 standard errors; the inflation factor is about 2.4x, exactly equal to sqrt(912/163).75% similarUnverifiedJoint adaptation raises the four-benchmark macro-average accuracy from 66.46% to 74.32%.74% similarUnverifiedWhen a 3B small verifier is introduced, ProVe reaches an average 2.7x throughput, but its average accuracy across six tasks is 2.6 percentage points lower than AR.71% similarUnverifiedUsing 4-bit quantization as a rough estimate, each parameter requires approximately 0.5 bytes on average70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1141678API
curl https://kongchang.com/api/v1/knowledge/claims/1141678MCP
get_claim(id=1141678)