Towards Comparable Knowledge Distillation in Semantic Image Segmentation

• URL: http://arxiv.org/abs/2309.03659v1
• Date: Thu, 7 Sep 2023 11:56:23 GMT
• Title: Towards Comparable Knowledge Distillation in Semantic Image Segmentation
• Abstract summary: Knowledge Distillation (KD) is one proposed solution to large model sizes and slow inference speed in semantic segmentation. We identify 25 proposed distillation loss terms from 14 publications in the last 4 years.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Knowledge Distillation (KD) is one proposed solution to large model sizes and slow inference speed in semantic segmentation. In our research we identify 25 proposed distillation loss terms from 14 publications in the last 4 years. Unfortunately, a comparison of terms based on published results is often impossible, because of differences in training configurations. A good illustration of this problem is the comparison of two publications from 2022. Using the same models and dataset, Structural and Statistical Texture Distillation (SSTKD) reports an increase of student mIoU of 4.54 and a final performance of 29.19, while Adaptive Perspective Distillation (APD) only improves student performance by 2.06 percentage points, but achieves a final performance of 39.25. The reason for such extreme differences is often a suboptimal choice of hyperparameters and a resulting underperformance of the student model used as reference point. In our work, we reveal problems of insufficient hyperparameter tuning by showing that distillation improvements of two widely accepted frameworks, SKD and IFVD, vanish when hyperparameters are optimized sufficiently. To improve comparability of future research in the field, we establish a solid baseline for three datasets and two student models and provide extensive information on hyperparameter tuning. We find that only two out of eight techniques can compete with our simple baseline on the ADE20K dataset.

Related papers

• Enhancing deep learning models for time series classification via knowledge distillation [0.49387059996366106]
  Knowledge Distillation (KD) addresses this by transferring knowledge from a large teacher model to a smaller, more efficient student model.<n>We evaluate our approach on UCR Archive, the largest benchmark repository of time series datasets.
  arXiv  Detail & Related papers  (2026-07-07T20:51:22Z)
• Evolving Knowledge Distillation for Lightweight Neural Machine Translation [1.53762673045917]
  Evolving Knowledge Distillation (EKD) is a progressive training framework in which the student model learns from a sequence of teachers with gradually increasing capacities.<n>EKD effectively bridges the capacity gap, enabling compact models to achieve performance close to that of much larger teacher models.
  arXiv  Detail & Related papers  (2026-05-11T03:21:24Z)
• Iterative Layer-wise Distillation for Efficient Compression of Large Language Models [0.42970700836450487]
  This work investigates distillation methods for large language models (LLMs) with the goal of developing compact models that preserve high performance.<n>An improved method based on the ShortGPT approach has been developed, building upon the idea of incorporating iterative evaluation of layer importance.<n> Experiments on the Qwen2.5-3B model show that the number of layers can be reduced from 36 to 28 (resulting in a 2.47 billion parameter model) with only a 9.7% quality loss, and to 24 layers with an 18% loss.
  arXiv  Detail & Related papers  (2025-11-07T09:00:26Z)
• Beyond Objects: Contextual Synthetic Data Generation for Fine-Grained Classification [31.116511358786084]
  Text-to-image (T2I) models are increasingly used for synthetic dataset generation.<n>Fine-tuning a T2I model with a few real examples can help improve the quality of synthetic training data.<n>We propose a fine-tuning strategy BOB (BeyondOBjects) to mitigate these concerns for fine-grained classification.
  arXiv  Detail & Related papers  (2025-10-28T05:40:14Z)
• OpenCodeReasoning: Advancing Data Distillation for Competitive Coding [61.15402517835137]
  We build a supervised fine-tuning (SFT) dataset to achieve state-of-the-art coding capability results in models of various sizes. Our models use only SFT to achieve 61.8% on LiveCodeBench and 24.6% on CodeContests, surpassing alternatives trained with reinforcement learning.
  arXiv  Detail & Related papers  (2025-04-02T17:50:31Z)
• Ultra-Resolution Adaptation with Ease [62.56434979517156]
  We propose a set of key guidelines for ultra-resolution adaptation termed emphURAE. We show that tuning minor components of the weight matrices outperforms widely-used low-rank adapters when synthetic data are unavailable. Experiments validate that URAE achieves comparable 2K-generation performance to state-of-the-art closed-source models like FLUX1.1 [Pro] Ultra with only 3K samples and 2K iterations.
  arXiv  Detail & Related papers  (2025-03-20T16:44:43Z)
• ALoRE: Efficient Visual Adaptation via Aggregating Low Rank Experts [71.91042186338163]
  ALoRE is a novel PETL method that reuses the hypercomplex parameterized space constructed by Kronecker product to Aggregate Low Rank Experts. Thanks to the artful design, ALoRE maintains negligible extra parameters and can be effortlessly merged into the frozen backbone.
  arXiv  Detail & Related papers  (2024-12-11T12:31:30Z)
• De-confounded Data-free Knowledge Distillation for Handling Distribution Shifts [32.1016787150064]
  Data-Free Knowledge Distillation (DFKD) is a promising task to train high-performance small models to enhance actual deployment without relying on the original training data. Existing methods commonly avoid relying on private data by utilizing synthetic or sampled data. This paper proposes a novel perspective with causal inference to disentangle the student models from the impact of such shifts.
  arXiv  Detail & Related papers  (2024-03-28T16:13:22Z)
• Augmentation-Free Dense Contrastive Knowledge Distillation for Efficient Semantic Segmentation [16.957139277317005]
  Augmentation-free Dense Contrastive Knowledge Distillation (Af-DCD) is a new contrastive distillation learning paradigm. Af-DCD trains compact and accurate deep neural networks for semantic segmentation applications.
  arXiv  Detail & Related papers  (2023-12-07T09:37:28Z)
• One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge Distillation [69.65734716679925]
  Knowledge distillation has proven to be a highly effective approach for enhancing model performance through a teacher-student training scheme. Most existing distillation methods are designed under the assumption that the teacher and student models belong to the same model family. We propose a simple yet effective one-for-all KD framework called OFA-KD, which significantly improves the distillation performance between heterogeneous architectures.
  arXiv  Detail & Related papers  (2023-10-30T11:13:02Z)
• Label-Aware Hyperbolic Embeddings for Fine-grained Emotion Classification [11.973115466838717]
  Fine-grained emotion classification (FEC) is a challenging task. FEC needs to handle subtle nuance between labels, which can be complex and confusing. Most existing models only address text classification problem in the euclidean space. We propose HypEmo, a novel framework that can integrate hyperbolic embeddings to improve the FEC task.
  arXiv  Detail & Related papers  (2023-06-26T16:28:33Z)
• DisWOT: Student Architecture Search for Distillation WithOut Training [0.0]
  We explore a novel training-free framework to search for the best student architectures for a given teacher. Our work first empirically show that the optimal model under vanilla training cannot be the winner in distillation. Our experiments on CIFAR, ImageNet and NAS-Bench-201 demonstrate that our technique achieves state-of-the-art results on different search spaces.
  arXiv  Detail & Related papers  (2023-03-28T01:58:45Z)
• Adversarial Dual-Student with Differentiable Spatial Warping for Semi-Supervised Semantic Segmentation [70.2166826794421]
  We propose a differentiable geometric warping to conduct unsupervised data augmentation. We also propose a novel adversarial dual-student framework to improve the Mean-Teacher. Our solution significantly improves the performance and state-of-the-art results are achieved on both datasets.
  arXiv  Detail & Related papers  (2022-03-05T17:36:17Z)
• DSEE: Dually Sparsity-embedded Efficient Tuning of Pre-trained Language Models [152.29364079385635]
  As pre-trained models grow bigger, the fine-tuning process can be time-consuming and computationally expensive. We propose a framework for resource- and parameter-efficient fine-tuning by leveraging the sparsity prior in both weight updates and the final model weights. Our proposed framework, dubbed Dually Sparsity-Embedded Efficient Tuning (DSEE), aims to achieve two key objectives: (i) parameter efficient fine-tuning and (ii) resource-efficient inference.
  arXiv  Detail & Related papers  (2021-10-30T03:29:47Z)
• Knowledge distillation: A good teacher is patient and consistent [71.14922743774864]
  There is a growing discrepancy in computer vision between large-scale models that achieve state-of-the-art performance and models that are affordable in practical applications. We identify certain implicit design choices, which may drastically affect the effectiveness of distillation. We obtain a state-of-the-art ResNet-50 model for ImageNet, which achieves 82.8% top-1 accuracy.
  arXiv  Detail & Related papers  (2021-06-09T17:20:40Z)
• When Liebig's Barrel Meets Facial Landmark Detection: A Practical Model [87.25037167380522]
  We propose a model that is accurate, robust, efficient, generalizable, and end-to-end trainable. In order to achieve a better accuracy, we propose two lightweight modules. DQInit dynamically initializes the queries of decoder from the inputs, enabling the model to achieve as good accuracy as the ones with multiple decoder layers. QAMem is designed to enhance the discriminative ability of queries on low-resolution feature maps by assigning separate memory values to each query rather than a shared one.
  arXiv  Detail & Related papers  (2021-05-27T13:51:42Z)
• MixKD: Towards Efficient Distillation of Large-scale Language Models [129.73786264834894]
  We propose MixKD, a data-agnostic distillation framework, to endow the resulting model with stronger generalization ability. We prove from a theoretical perspective that under reasonable conditions MixKD gives rise to a smaller gap between the error and the empirical error. Experiments under a limited-data setting and ablation studies further demonstrate the advantages of the proposed approach.
  arXiv  Detail & Related papers  (2020-11-01T18:47:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.