Unverified50% confidenceBenchmarkExact time
On point-in-box accuracy, the 279M ranker reaches 73.4%, zero-shot UI-TARS-2B reaches 29.6%, and Qwen2.5-VL-7B reaches 23.7%.
1
Sources
50%
Confidence
Long-term
Relevance
11/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn "continuation of previous operation" type utterances, Qwen3.6-35B-A3B achieved an accuracy of 86.3%, while the 4B and 8B models achieved only 70.4%.74% similarUnverifiedThe Top3 configuration achieved an accuracy of 97.148%, a weighted F1 of 97.055%, and an MCC of 0.9675, comprehensively outperforming the Full21 configuration using all 21 metrics (accuracy 95.999%, weighted F1 95.521%, MCC 0.9549).72% similarUnverifiedQ64's 74.32% accuracy is still below the native 27B baseline's 75.68%, and also below AnyJev Qwen3-32B L0's 76.68%.71% similarUnverifiedWhen fine-tuning Qwen on a single RTX 3090, the number of correct answers on the dev set rose from 7/28 to 13/28, but the average negative log-likelihood (NLL) worsened from 2.215 to 3.916.71% similarUnverifiedThe dense Qwen3.8-27B model's 3-bit version actually scored 0.05 higher (within noise range), but the number of tokens generated per task increased by 40% (confidence interval 1.25–1.59x).71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1178502API
curl https://kongchang.com/api/v1/knowledge/claims/1178502MCP
get_claim(id=1178502)