Decision-calibrated conformal uncertainty for pacing uses the support function of the signed policy sensitivity set to achieve smaller uncertainty radii on public datasets.
InProceedings of the 1st Workshop on Deep Learning for Recommender Systems(Boston, MA, USA) (DLRS 2016)
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
DREAM proposes intent-aware tokenization, frozen-model evaluation, and dynamic beams to refine early SID assignments and improve cold-start performance in generative recommenders on Amazon benchmarks.
LeAP is a model-agnostic plug-in that learns to permute and rank heterogeneous sparse features for selection, achieving SOTA on public datasets and removing over 3600 redundant features in a production system with 12k features.
FLUID introduces LUCID semantic codes from a multimodal encoder to retire item IDs in livestreaming rankers, with staged warmup yielding online gains of +0.55% watch duration and +2.05% cold-start views.
SAILRec uses dual-side semantic alignment and hierarchical attention steering to improve how LLMs incorporate collaborative embeddings for recommendations, outperforming baselines on MovieLens-1M and Amazon-Book datasets.
citing papers explorer
-
FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation
FLUID introduces LUCID semantic codes from a multimodal encoder to retire item IDs in livestreaming rankers, with staged warmup yielding online gains of +0.55% watch duration and +2.05% cold-start views.