Unverified50% confidenceFactExact time
The tensor PCA problem is defined on the sphere S_N: given noisy observations of the k-th order tensor power of an unknown vector (k≥3 fixed), the goal is to recover that unknown vector.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTensor PCA exhibits an information-computation gap: when the signal-to-noise ratio is above the algorithmic threshold, global algorithms such as spectral methods and sum-of-squares can successfully recover in polynomial time; below that threshold, lower bounds for these algorithms and for certain restricted computational models show that they must fail.71% similarUnverifiedThe VRAM requirement estimation formula is: VRAM requirement ≈ weight size + KV cache + inference intermediate activation overhead; the weight part is parameter count × bytes per parameter, and the KV cache part is 2 × number of layers × hidden dimension × context length × bytes per element.62% similarUnverifiedMarchenko-Pastur定律描述了p>>n高维稀疏条件下随机矩阵特征值的分布规律，揭示许多看似显著的主成分可能仅反映噪声结构61% similarUnverifiedIn vector retrieval, a top-k value that is too small may miss information, while one that is too large introduces noise and crowds the context window, so it needs tuning according to document granularity and question type.59% similarUnverifiedPRSR's low-rank subspace estimation applies SVD or PCA to the matrix formed by the residual vectors of parallel sentence pairs, takes the top k singular vectors to span the subspace, and projects the final-layer token hidden states onto the orthogonal complement of that subspace.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114779API
curl https://kongchang.com/api/v1/knowledge/claims/1114779MCP
get_claim(id=1114779)