Unverified50% confidenceBenchmarkExact time
Testing the query prefix mentioned in the bge model documentation, "Represent this sentence for searching relevant passages:", the top score improved by 0.014, but the gap to the second place actually narrowed.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/17/2026
First Seen
Valid until: 10/31/2026
Sources
Related Entities
Related Claims
UnverifiedThe limitations of this test include: limited task scale (50 tasks, 3 trials per prompt), style versions written by a single author, length confounding between the florid and normal versions, testing only English and local models below 14B, and the perplexity check covering only three smaller Qwen models and using full-precision weights while actual inference used 4-bit quantization.61% similarUnverified本次评测使用全国一卷高考数学题对比Muse Glimmer和通义千问3.6 27B59% similarUnverifiedThe Laya decision model responds up to 4.1 times faster for short requests, with a 3-question guardrail check on a B200 dropping from 11.6 ms to 2.8 ms.59% similarUnverifiedThe format instruction on the last line of the prompt is the key variable determining the output quality of small models: on math tasks, after unifying the last-line format, differences in question text style caused at most a 5-point fluctuation, while changing the format line could produce a 45-point gap.59% similarUnverifiedCaveman's fixed benchmark test results from 54 runs show a 33.2% reduction in input tokens, with all 18/18 correctness checks passing.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1054552API
curl https://kongchang.com/api/v1/knowledge/claims/1054552MCP
get_claim(id=1054552)