BAHSD applies adaptive hierarchical distillation with dynamic-temperature KL divergence, ranking consistency, and InfoNCE contrastive learning to improve black-box sequential recommendation extraction on long-tail sequences.
arXiv preprint arXiv:2505.04560 (2025)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
BMLR reshapes the cross-modal label space to equalize mapping difficulty and balance optimization across modalities in multimodal learning.
citing papers explorer
-
BAHSD: Bridging the Long-tail Gap via Adaptive Distillation in Black-box Sequential Recommendation
BAHSD applies adaptive hierarchical distillation with dynamic-temperature KL divergence, ranking consistency, and InfoNCE contrastive learning to improve black-box sequential recommendation extraction on long-tail sequences.
-
Balancing Multimodal Learning through Label Space Reshaping
BMLR reshapes the cross-modal label space to equalize mapping difficulty and balance optimization across modalities in multimodal learning.