Unverified50% confidenceFactExact time
Appendix B.1 of the paper describes the problem of small errors accumulating progressively over long latent-space segments, gradually eroding the semantic coherence of the latent trajectory.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper points out that the root cause of long-context degradation lies not entirely in context length, but in the number of effective distractors it contains.68% similarUnverifiedThe paper covers two settings, compactly supported targets and power-law targets, under the p-Wasserstein distance framework, gives non-asymptotic convergence bounds, and separates the contributions of early-stopping error, initialization error, score approximation error, and time discretization error.68% similarUnverifiedLong prompts suffer from execution decay: the more constraints there are, the greater the deviation between actual output and textual description, and a more detailed prompt does not equal more accurate model execution.67% similarUnverifiedDecoding from continuous representations to discrete tokens is an error-prone step; minor deviations in the embedding space can lead to decoding errors, and quantization errors are amplified as sequence length grows.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1069599API
curl https://kongchang.com/api/v1/knowledge/claims/1069599MCP
get_claim(id=1069599)