Unverified50% confidenceFactExact time
Across 9,000 greedy-decoding questions covering all three splits, token ID 3582 (corresponding to the bare word 'yes') and 1217 (corresponding to the bare word 'no') were never once generated by LLaVA-1.5-7B; what the model actually outputs is ' Yes' (token 3869) and ' No' (token 1939).
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe noul type returns a numerical value with a confidence score; the name is a fusion of no/null/yes, representing a continuous three-state semantic space: close to 0 indicates negation, close to 0.5 indicates uncertainty, and close to 1 indicates affirmation.63% similarUnverifiedThe 4-bit LLaMA 3.1 8B dropped from 67.3 to 67.1 on the HumanEval coding benchmark.60% similarUnverifiedAdding 3 neutral paraphrased versions per cell beyond the original prompt, 0 out of 112 configurations qualified (0/112).60% similarUnverifiedIn EvoMap's benchmark test, sub-agents actually found 373 correct answers, but after the main agent merged the results, only 217 were retained in the final output.59% similarUnverifiedThe 8-bit signed number -5, reinterpreted as an unsigned number, is 251, i.e., 256 + (-5).58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074064API
curl https://kongchang.com/api/v1/knowledge/claims/1074064MCP
get_claim(id=1074064)