A jointly learned hierarchical index with cross-attention and residual quantization scales exact retrieval in foundational recommendation models, deployed at Meta with additional performance from test-time training on index nodes.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 3verdicts
UNVERDICTED 3roles
background 1polarities
background 1representative citing papers
The Isotonic Layer unifies calibration and debiasing in recommendation models as a single end-to-end trainable component using learnable context embeddings for piecewise linear adjustments.
Multimodal fusion of MLLM-generated text embeddings and visual features improves retrieval for forensic tattoo and face matching tasks across images, descriptions, and sketches.
citing papers explorer
-
Efficient Retrieval Scaling with Hierarchical Indexing for Large Scale Recommendation
A jointly learned hierarchical index with cross-attention and residual quantization scales exact retrieval in foundational recommendation models, deployed at Meta with additional performance from test-time training on index nodes.
-
Isotonic Layer: A Unified Framework for Recommendation Calibration and Debiasing
The Isotonic Layer unifies calibration and debiasing in recommendation models as a single end-to-end trainable component using learnable context embeddings for piecewise linear adjustments.
-
Bridging the Modality Gap in Forensic Image Retrieval
Multimodal fusion of MLLM-generated text embeddings and visual features improves retrieval for forensic tattoo and face matching tasks across images, descriptions, and sketches.