Unverified50% confidenceFactExact time
In retrieval tasks, when the context is extended with neutral filler text, the retrieval accuracy of five open-source models (0.5B to 3B parameters) is almost unaffected, staying between 0.97 and 1.00.
1
Sources
50%
Confidence
Long-term
Relevance
11/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLexical coverage measurements showed: exact match 1.00, paraphrase 0.44, third-party question 0.63; with the threshold set at 0.9, queries that genuinely need semantics open the semantic channel.70% similarUnverifiedThe 0.6B version can compete with models several times its parameter count on visual document retrieval tasks.69% similarUnverifiedIn DeepSeek's official comparison, when running with the same checkpoint at best effort, its minimal framework scored 72.6 while the open-source code scored 65.5.67% similarUnverifiedOn the 35 prompts that the router kept local, Llama 3.2 1B's answers were judged acceptable 30 times (86%, average 4.4/5); on the 23 text prompts sent to the cloud, the same model was acceptable only 4 times (17%, average 2.0/5).66% similarUnverifiedcheckwash 0.6.0 achieved a precision of 0.67 and recall of 0.22 for its own block judgment on this corpus, while the raw count signal read independently yielded a precision of 0.80 and recall of 0.89.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134059API
curl https://kongchang.com/api/v1/knowledge/claims/1134059MCP
get_claim(id=1134059)