A trunk-branch method for adapting to conflicting time series tasks reports large error reductions on a synthetic benchmark, but the comparison is confounded by unequal adaptation budgets and the theory overclaims relative to LoRA.
By definition, the dynamic hypothesis space is the union over time: Hdyn = [ t∈[0,T] Vt = K[ i=1 Fi, whereFi :=S t∈[ti−1,ti)Vt denotes the function class active in segmenti
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Dynamic Perturbed Adaptive Method for Infinite Task-Conflicting Time Series
A trunk-branch method for adapting to conflicting time series tasks reports large error reductions on a synthetic benchmark, but the comparison is confounded by unequal adaptation budgets and the theory overclaims relative to LoRA.