Unverified50% confidenceBenchmarkExact time
After fine-tuning Qwen3-14B with LoRA SFT under five-fold cross-validation, the model reached κ = 0.524, close to the inter-coder human ceiling of κ = 0.527.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/3/2026
First Seen
Valid until: 2/1/2027
Sources
Related Entities
Related Claims
UnverifiedUsing the Qwen3-27B model on the wikitext-2 dataset with fp32 reference logits as a baseline for KL divergence measurement, the median KLD dropped from 0.0023 to 0.000001, an approximately 2300-fold improvement in precision.73% similarUnverifiedThe self-hosted Qwen3.6-27B model, used as a replacement judge, achieved an agreement level of kappa=0.7268% similarUnverifiedFor the Qwen2.5-1.5B model, the int4 theoretical roofline ceiling for batch-1 decoding is 411.8 tok/s, while the actual self-developed kernel reaches only 25.5 tok/s, a gap of about 16x.67% similarUnverifiedOn BU Bench V1, the 4B-parameter SFT model surpassed the 45% score reported by DeepSeek V4 Pro and Kimi K2.6 under the open-source Browser Use framework.67% similarUnverifiedOn the GSM8K mathematical reasoning test, Qwen3-4B scores 0.915 and Gemma 4 E2B scores 0.911, a difference of +0.4 percentage points with a confidence interval crossing zero, making it the only model on par with Gemma.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1129268API
curl https://kongchang.com/api/v1/knowledge/claims/1129268MCP
get_claim(id=1129268)