# Nested Hash Layer: A Plug-and-play Module for Multiple-length Hash Code Learning

arXiv: https://arxiv.org/abs/2412.08922 · HTML: https://mlsanity.com/paper/2412.08922/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 49 references that are in the map.
- First posted on arXiv 12 Dec 2024.

## Builds on (most cited first, 26 of 26)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [End-to-End Multi-Task Learning with Attention](https://mlsanity.com/paper/1803.10704/) (2018)
- [Conflict-Averse Gradient Descent for Multi-task Learning](https://mlsanity.com/paper/2110.14048/) (2021)
- [Matryoshka Representation Learning](https://mlsanity.com/paper/2205.13147/) (2022)
- [Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout](https://mlsanity.com/paper/2010.06808/) (2020)
- [UberNet: Training a `Universal' Convolutional Neural Network for Low-, Mid-, and High-Level Vision using Diverse Datasets and Limited Memory](https://mlsanity.com/paper/1609.02132/) (2016)
- [RotoGrad: Gradient Homogenization in Multitask Learning](https://mlsanity.com/paper/2103.02631/) (2021)
- [Central Similarity Quantization for Efficient Image and Video Retrieval](https://mlsanity.com/paper/1908.00347/) (2019)
- [Recon: Reducing Conflicting Gradients from the Root for Multi-Task Learning](https://mlsanity.com/paper/2302.11289/) (2023)
- [MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning](https://mlsanity.com/paper/2003.14058/) (2020)
- [A Survey on Deep Hashing Methods](https://mlsanity.com/paper/2003.03369/) (2020)
- [One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective](https://mlsanity.com/paper/2109.14449/) (2021)
- [Stochastic Filter Groups for Multi-Task CNNs: Learning Specialist and Generalist Convolution Kernels](https://mlsanity.com/paper/1908.09597/) (2019)
- [Asymmetric Deep Supervised Hashing](https://mlsanity.com/paper/1707.08325/) (2017)
- [Deep Supervised Hashing with Triplet Labels](https://mlsanity.com/paper/1612.03900/) (2016)
- [Unsupervised Hashing with Contrastive Information Bottleneck](https://mlsanity.com/paper/2105.06138/) (2021)
- [Deep Triplet Quantization](https://mlsanity.com/paper/1902.00153/) (2019)
- [Hashing with Mutual Information](https://mlsanity.com/paper/1803.00974/) (2018)
- [Collaborative Learning for Extremely Low Bit Asymmetric Hashing](https://mlsanity.com/paper/1809.09329/) (2018)
- [Push for Quantization: Deep Fisher Hashing](https://mlsanity.com/paper/1909.00206/) (2019)

## Built on by (most cited first, 2 of 2)

- [UniHash: Unifying Pointwise and Pairwise Hashing Paradigms](https://mlsanity.com/paper/2601.09828/) (2026)
- [Matryoshka Hash Representations for Model-Aware Compact Semantic Retrieval](https://mlsanity.com/paper/2609.07276/) (2026)

## Abstract (arXiv)

Deep supervised hashing is essential for efficient storage and search in large-scale image retrieval. Traditional deep supervised hashing models generate single-length hash codes, but this creates a trade-off between efficiency and effectiveness for different code lengths. To find the optimal length for a task, multiple models must be trained, increasing time and computation. Furthermore, relationships between hash codes of different lengths are often ignored. To address these issues, we propose the Nested Hash Layer (NHL), a plug-and-play module for deep supervised hashing models. NHL generates hash codes of multiple lengths simultaneously in a nested structure. To resolve optimization conflicts from multiple learning objectives, we introduce a dominance-aware dynamic weighting strategy to adjust gradients. Additionally, we propose a long-short cascade self-distillation method, where long hash codes guide the learning of shorter ones, improving overall code quality. Experiments indicate that the NHL achieves an overall training speed improvement of approximately 5 to 8 times across various deep supervised hashing models and enhances the average performance of these models by about 3.4%.