Unverified50% confidenceFactExact time
Multi-hop transitivity: if dataset α pairs (B₁, B₂) and dataset β pairs (B₂, B₃), then fine-tuning only on B₁ can elicit B₃, even if B₁ and B₃ never co-occur in a single document.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUsing a 4B model to rerank the candidate shortlist filtered by BM25 outperformed BM25 alone (+0.047, confidence interval [0.020, 0.073]), even though the 4B reranker itself did not pass qualification.61% similarUnverifiedThe conversion from FP8 to BF16 is lossless because BF16 can exactly represent all values representable by FP8 E4M3, while the reverse is not true59% similarUnverifiedVector retrieval returns cosine similarity or distance scores, while BM25 returns term-frequency-based scores; the two have different units, so raw scores cannot be directly added or compared.58% similarUnverifiedThe 3.8-27B has fewer parameters than the 3.5/3.6-35B-A3B series.57% similarUnverifiedRegardless of the encoding scheme used, 4-bit precision can only distinguish 16 different numerical levels.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1070460API
curl https://kongchang.com/api/v1/knowledge/claims/1070460MCP
get_claim(id=1070460)