Pretraining Multiple Instance Learning Networks with Multi-Teacher Distillation from Pathology Slide Foundation Models

• URL: http://arxiv.org/abs/2607.14703v1
• Date: Thu, 16 Jul 2026 08:04:25 GMT
• Title: Pretraining Multiple Instance Learning Networks with Multi-Teacher Distillation from Pathology Slide Foundation Models
• Abstract summary: Multiple instance learning (MIL) has become the main paradigm for whole-slide image (WSI) analysis in computational pathology.<n>Existing MIL aggregators are still typically trained from scratch for each downstream task.<n>We propose a distillation-based pretraining framework for MIL.
• Score: 10.354992690532146
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multiple instance learning (MIL) has become the main paradigm for whole-slide image (WSI) analysis in computational pathology. However, existing MIL aggregators are still typically trained from scratch for each downstream task, relying on limited slide-level labels to learn both aggregation mechanisms and downstream discriminative representations simultaneously. As a result, they often suffer from unstable optimization, overfitting, and limited transferability. Similar to pretrained ResNet and Vision Transformer models in natural image learning, MIL also requires reusable pretrained initialization. However, high-quality slide-level pretraining data remain scarce, and MIL models are usually lightweight and weakly supervised, making large-scale pretraining difficult in practice. To address this challenge, we propose a distillation-based pretraining framework for MIL, which leverages two slide-level foundation models, TITAN and CARE, as teachers to transfer their representational knowledge into a diverse set of MIL architectures. To effectively balance supervision from different teachers, we further introduce an angular dispersion normalized distillation loss. The distilled weights are then used as initialization for downstream adaptation. We conduct systematic evaluations on 15 benchmark datasets under both linear probing and full-parameter fine-tuning, and further validate its advantages in few-shot scenarios. Experimental results show that pretraining generally improves MIL aggregators over from scratch training, especially in linear-probing and few-shot settings, while maintaining the computational efficiency of lightweight MIL models. Code is available at https://github.com/fu0201/MIL_Pretrained.

Related papers

• Do Multiple Instance Learning Models Transfer? [7.766368937536295]
  Multiple Instance Learning (MIL) is a cornerstone approach in computational pathology (CPath)<n>MIL often struggles with small, weakly supervised clinical datasets.<n>In this study, we evaluate the transfer learning capabilities of pretrained MIL models.
  arXiv  Detail & Related papers  (2025-06-10T17:50:04Z)
• SimMIL: A Universal Weakly Supervised Pre-Training Framework for Multi-Instance Learning in Whole Slide Pathology Images [12.827931905880163]
  This paper proposes to pre-train feature extractor for MIL via a weakly-supervised scheme.<n>To learn effective features for MIL, we delve into several key components, including strong data augmentation, a non-linear prediction head and the robust loss function.<n>We conduct experiments on common large-scale WSI datasets and find it achieves better performance than other pre-training schemes.
  arXiv  Detail & Related papers  (2025-05-10T17:23:36Z)
• RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models [53.571195477043496]
  We propose an algorithm named Rotated Straight-Through-Estimator (RoSTE)<n>RoSTE combines quantization-aware supervised fine-tuning (QA-SFT) with an adaptive rotation strategy to reduce activation outliers.<n>Our findings reveal that the prediction error is directly proportional to the quantization error of the converged weights, which can be effectively managed through an optimized rotation configuration.
  arXiv  Detail & Related papers  (2025-02-13T06:44:33Z)
• Rethinking Pre-Trained Feature Extractor Selection in Multiple Instance Learning for Whole Slide Image Classification [2.375943263571389]
  Multiple instance learning (MIL) has become a preferred method for gigapixel whole slide image (WSI) classification without requiring patch-level annotations.<n>This study systematically evaluating MIL feature extractors across three dimensions: pre-training dataset, backbone model, and pre-training method.
  arXiv  Detail & Related papers  (2024-08-02T10:34:23Z)
• Diffusion-Based Neural Network Weights Generation [80.89706112736353]
  D2NWG is a diffusion-based neural network weights generation technique that efficiently produces high-performing weights for transfer learning. Our method extends generative hyper-representation learning to recast the latent diffusion paradigm for neural network weights generation. Our approach is scalable to large architectures such as large language models (LLMs), overcoming the limitations of current parameter generation techniques.
  arXiv  Detail & Related papers  (2024-02-28T08:34:23Z)
• An Emulator for Fine-Tuning Large Language Models using Small Language Models [91.02498576056057]
  We introduce emulated fine-tuning (EFT), a principled and practical method for sampling from a distribution that approximates the result of pre-training and fine-tuning at different scales. We show that EFT enables test-time adjustment of competing behavioral traits like helpfulness and harmlessness without additional training. Finally, a special case of emulated fine-tuning, which we call LM up-scaling, avoids resource-intensive fine-tuning of large pre-trained models by ensembling them with small fine-tuned models.
  arXiv  Detail & Related papers  (2023-10-19T17:57:16Z)
• Towards Foundation Models for Scientific Machine Learning: Characterizing Scaling and Transfer Behavior [32.74388989649232]
  We study how pre-training could be used for scientific machine learning (SciML) applications. We find that fine-tuning these models yields more performance gains as model size increases.
  arXiv  Detail & Related papers  (2023-06-01T00:32:59Z)
• Black Box Few-Shot Adaptation for Vision-Language models [41.49584259596654]
  Vision-Language (V-L) models trained with contrastive learning to align the visual and language modalities have been shown to be strong few-shot learners. We describe a black-box method for V-L few-shot adaptation that operates on pre-computed image and text features. We propose Linear Feature Alignment (LFA), a simple linear approach for V-L re-alignment in the target domain.
  arXiv  Detail & Related papers  (2023-04-04T12:42:29Z)
• eP-ALM: Efficient Perceptual Augmentation of Language Models [70.47962271121389]
  We propose to direct effort to efficient adaptations of existing models, and propose to augment Language Models with perception. Existing approaches for adapting pretrained models for vision-language tasks still rely on several key components that hinder their efficiency. We show that by freezing more than 99% of total parameters, training only one linear projection layer, and prepending only one trainable token, our approach (dubbed eP-ALM) significantly outperforms other baselines on VQA and Captioning.
  arXiv  Detail & Related papers  (2023-03-20T19:20:34Z)
• SPDF: Sparse Pre-training and Dense Fine-tuning for Large Language Models [4.114555639014612]
  We show the benefits of using unstructured weight sparsity to train only a subset of weights during pre-training. We demonstrate that we can induce up to 75% sparsity into a 1.3B parameter GPT-3 XL model resulting in a 2.5x reduction in pre-training FLOPs.
  arXiv  Detail & Related papers  (2023-03-18T17:56:01Z)
• Slimmable Networks for Contrastive Self-supervised Learning [69.9454691873866]
  Self-supervised learning makes significant progress in pre-training large models, but struggles with small models. We introduce another one-stage solution to obtain pre-trained small models without the need for extra teachers. A slimmable network consists of a full network and several weight-sharing sub-networks, which can be pre-trained once to obtain various networks.
  arXiv  Detail & Related papers  (2022-09-30T15:15:05Z)
• Model-Agnostic Multitask Fine-tuning for Few-shot Vision-Language Transfer Learning [59.38343286807997]
  We propose Model-Agnostic Multitask Fine-tuning (MAMF) for vision-language models on unseen tasks. Compared with model-agnostic meta-learning (MAML), MAMF discards the bi-level optimization and uses only first-order gradients. We show that MAMF consistently outperforms the classical fine-tuning method for few-shot transfer learning on five benchmark datasets.
  arXiv  Detail & Related papers  (2022-03-09T17:26:53Z)
• Dual-stream Maximum Self-attention Multi-instance Learning [11.685285490589981]
  Multi-instance learning (MIL) is a form of weakly supervised learning where a single class label is assigned to a bag of instances while the instance-level labels are not available. We propose a dual-stream maximum self-attention MIL model (DSMIL) parameterized by neural networks. Our method achieves superior performance compared to the best MIL methods and demonstrates state-of-the-art performance on benchmark MIL datasets.
  arXiv  Detail & Related papers  (2020-06-09T22:44:58Z)

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.