Unverified50% confidenceBenchmarkExact time
Counterfactual mDPO improved METEOR from 58.10% to 85.30%, reduced Self-BLEU-2 from 58.79% to 52.88%, and improved Distinct-3 from 6.68% to 7.81%.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCompared to standard LoRA, counterfactual mDPO improved decision accuracy from 93.33% to 96.67%.70% similarUnverifiedRaising the confidence level from 95% to 99% (z-value 2.576) will expand the margin of error by approximately 31% for the same sample size.68% similarUnverifiedEvidence-driven pre-improvement validation improved precision from 58.6% to 72.3%, and reduced unsupported objectives per round from 1.93 to 0.50.64% similarUnverifiedOn GSM8K, Flash-Next scored 97.5% versus FP8's 90.0%; on German MGSM, 92.5% versus 90.0%; and on IFEval, both tied at 89.6%.63% similarUnverifiedIn the control group where the answer key was removed from the prompt, the match rate plummeted from 63% to 1% (1/75).63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975672API
curl https://kongchang.com/api/v1/knowledge/claims/975672MCP
get_claim(id=975672)