AsymRec decouples input and output representations in generative recommendation via multi-expert semantic projection and multi-faceted hierarchical quantization, outperforming prior models by 15.8% on average.
Stacked Quantizers for Compositional Vector Compression
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
Recently, Babenko and Lempitsky introduced Additive Quantization (AQ), a generalization of Product Quantization (PQ) where a non-independent set of codebooks is used to compress vectors into small binary codes. Unfortunately, under this scheme encoding cannot be done independently in each codebook, and optimal encoding is an NP-hard problem. In this paper, we observe that PQ and AQ are both compositional quantizers that lie on the extremes of the codebook dependence-independence assumption, and explore an intermediate approach that exploits a hierarchical structure in the codebooks. This results in a method that achieves quantization error on par with or lower than AQ, while being several orders of magnitude faster. We perform a complexity analysis of PQ, AQ and our method, and evaluate our approach on standard benchmarks of SIFT and GIST descriptors, as well as on new datasets of features obtained from state-of-the-art convolutional neural networks.
years
2026 3verdicts
UNVERDICTED 3representative citing papers
GSRQ applies a gain-shape variant of K-means inside residual quantization to improve directional fidelity, raising LongBench accuracy from 11.34 to 33.54 at 1-bit on LLaMA-3-8B.
DPPMG learns discrete modal-specific preferences via a dedicated GNN from multimodal user data, quantizes them into tokens, and feeds them into generators with a consistency reward to produce personalized text and images.
citing papers explorer
-
Asymmetric Generative Recommendation via Multi-Expert Projection and Multi-Faceted Hierarchical Quantization
AsymRec decouples input and output representations in generative recommendation via multi-expert semantic projection and multi-faceted hierarchical quantization, outperforming prior models by 15.8% on average.
-
GSRQ: Gain-Shape Residual Quantization for Sub-1-bit KV Cache
GSRQ applies a gain-shape variant of K-means inside residual quantization to improve directional fidelity, raising LongBench accuracy from 11.34 to 33.54 at 1-bit on LLaMA-3-8B.
-
Discrete Preference Learning for Personalized Multimodal Generation
DPPMG learns discrete modal-specific preferences via a dedicated GNN from multimodal user data, quantizes them into tokens, and feeds them into generators with a consistency reward to produce personalized text and images.