Selective Prediction Forgetting for Adaptation of Weakly Supervised Localization of Histology Images

Alexis Guichemerre; Banafsheh Karimian; Soufiane Belharbi; Natacha Gillet; Nicolas Thome; Pourya Shamsolmoali; Mohammadhadi Shateri; Luke McCaffrey

cs.CVMar 12, 2026

Selective Prediction Forgetting for Adaptation of Weakly Supervised Localization of Histology Images

Authors: Alexis Guichemerre, Banafsheh Karimian, Soufiane Belharbi, Natacha Gillet, Nicolas Thome, Pourya Shamsolmoali, Mohammadhadi Shateri, Luke McCaffrey,

Organizations: ÉTS Montréal, Montreal, Canada · Sorbonne Université, Paris, France · University of York, York, United Kingdom · McGill University, Montreal, Canada

Abstract

Under strong cross-domain shifts, weakly supervised object localization (WSOL) models often lose their capacity to discriminate and develop a strong prediction bias toward a subset of classes. As a result, the majority of target samples are assigned to these dominant classes. Since source-free (unsupervised) domain adaptation (SFDA) methods typically rely on pseudo-labels, inaccurate pseudo-label estimation causes suboptimal adaptation and poor classification and localization performance. This issue is particularly challenging in computational histopathology, where shifts in organ-specific morphology, staining protocols, and acquisition conditions can significantly impact class-discriminative visual cues. This paper introduces SFDA with selective prediction forgetting (SFDA-SPF), a method designed to mitigate excessive concentration of predictions for dominant classes during target-domain adaptation. SFDA-SPF identifies prediction-dominant classes based on the model's target predictions and selectively revises uncertain samples assigned to them. Importantly, this procedure does not require knowledge of the true target class distribution. In contrast with previous SFDA methods, SFDA-SPF introduces a 'forget set' composed of uncertain samples from dominant classes near their decision boundaries, enabling the classification boundaries to shift away from biased predictions. A second 'retain set' is used to anchor certain and accurate samples. To forget the predictions of dominant classes, SFDA-SPF introduces a complementary optimization objective that reduces their probability while allowing alternative class assignments to emerge. This iterative optimization process reassesses target images for forget and retain sets at regular intervals. In addition to improving image-level pseudo-label estimation, the pixel-level classifier is jointly optimized, further enhancing localization.

Figures & tables

Figure 1: (a) SFDA-SPF reduces predictive bias across centers. (b) Source-only inference shows a strong predictive bias toward the cancer class when transferring from GlaS (source) to CAMELYON16 and CAMELYON17 (targets) with PixelCAM. (c) In contrast, traditional SFDA methods, such as SFDA-DE , amplify the bias and often collapse toward a dominant class.

Figure 2: Conceptual overview of SFDA-SPF . In source-free adaptation for histological WSOL, the source model is biased toward the dominant class. (a) Source model: biased prediction. Many samples lie near the decision boundary, where high-entropy predictions are informative but error-prone. (b) Standard SFDA: trust pseudo-labels. Conventional methods reinforce confident pseudo-labels. However, in histology, confidence does not necessarily imply correctness, and erroneous pseudo-labels amplify the initial bias and attract uncertain samples toward the dominant class. (c) SFDA-SPF forget uncertain samples (ours). We instead identify uncertain dominant-class samples as a forget set . The forgetting objective reduces the probability assigned to the current dominant-class prediction, allowing probability mass to be redistributed toward alternative classes.

Figure 3: Overview of the proposed SFDA-SPF pipeline. Starting from a source-pretrained WSOL model and an unlabeled target set, SFDA-SPF first identifies dominant classes from the target prediction distribution. For each dominant class, the fraction $\rho$ of samples with the highest normalized entropy is assigned to the forget set, while the remaining target samples form the retain set. During adaptation, $\mathcal{L}_{\mathrm{F}}$ reduces the probability assigned to the current dominant-class predictions of forget samples, whereas $\mathcal{L}_{\mathrm{R}}$ preserves the current assignments of the remaining samples to limit excessive model drift. An auxiliary pixel-level loss, $\mathcal{L}_{\mathrm{L}}$ , uses CAM-based partial pseudo-labels from the frozen source model to preserve spatial representations. Target predictions and the dominant, forget, and retain sets are periodically recomputed as the decision boundary evolves.

		C16		C17-0		C17-1		C17-2		C17-3		C17-4		Avg
	Method	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL
PixelCAM (MIDL’25)	Source only	37.9	51.8	37.2	41.2	27.2	58.7	44.9	64.9	43.1	39.1	31.1	39.9	36.9	49.3
	SHOT (ICML’20)	14.9	67.1	12.1	71.7	19.2	70.9	24.4	71.6	15.1	78.9	13.9	56.4	16.6	69.4
	SFDA-DE (CVPR’22)	37.9	51.8	14.5	50.0	36.7	65.7	28.9	70.5	18.9	49.9	31.1	39.9	28.0	54.6
	CDCL (TMM’22)	37.9	51.8	25.3	50.0	21.2	59.3	25.5	69.9	19.3	49.9	22.6	40.9	25.3	53.6
	ERL (ICLR’23)	29.9	59.7	13.2	59.5	14.6	67.4	43.0	69.4	19.3	49.9	32.4	54.0	25.4	60.0

Table 1: Localization ( PxAP ) and classification ( CL ) accuracy. Results shown for GlaS → C16 , C17-0 , C17-1 , C17-2 , C17-3 , C17-4 .

		C17-1		C17-2		C17-3		C17-4
	Methods	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL
PixelCAM (MIDL’25)	SHOT (ICML’20)	22.9	52.9	31.6	77.0	46.8	88.5	51.4	74.1
	SFDA-DE (CVPR’22)	23.9	49.4	18.7	50.0	50.7	92.0	51.4	74.1
	CDCL (TMM’22)	40.1	73.8	30.6	74.7	59.4	88.3	50.2	83.2
	ERL (ICLR’23)	17.4	62.2	45.6	58.4	50.7	92.0	51.4	74.1
	RGV (CVPR’25)	36.1	56.9	41.2	50.00	50.7	92.0	54.3	70.4

Table 2: Localization ( PxAP ) and classification ( CL ) accuracy. Inter-center SFDA on CAMELYON17 . The source model is trained on center 0 and independently adapted to centers 1–4.

		$50{:}50$		$60{:}40$		$70{:}30$		$80{:}20$		$90{:}10$
	Methods	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL	PxAP	CL
PixelCAM (MIDL’25)	SHOT (ICML’20)	19.2	70.0	22.9	41.6	37.2	41.2	24.2	41.6	30.5	73.6
	CDCL (TMM’22)	25.3	50.0	25.5	45.4	22.6	44.6	22.4	44.6	22.8	44.6
	ERL (ICLR’23)	13.2	59.5	21.8	44.2	20.4	43.1	21.5	43.5	21.0	43.5
	RGV (CVPR’25)	31.0	50.0	33.7	50.0	33.3	50.0	33.1	50.0	32.8	50.0
	Ours	49.9	86.2	43.7	79.0	55.1	82.8	46.0	79.4	58.8	82.8

Table 3: Robustness to class imbalance in the target adaptation set on CAMELYON17 . Starting from the balanced setting (

50{:}50

), we progressively increase the proportion of Normal samples up to

90{:}10

. The test set remains balanced across all settings; only the target training set is imbalanced.

Panel of Figure 4 — Figure 4: Effect of dynamic resampling frequency during adaptation with PixelCAM .

$\mathcal{L}_{\mathrm{R}}$	$\mathcal{L}_{\mathrm{F}}$	$\mathcal{L}_{\mathrm{L}}$	PxAP $\uparrow$	CL $\uparrow$
✓			27.6	41.2
	✓		10.4	85.1
		✓	9.5	48.1
✓	✓		13.1	84.7
✓	✓	✓	49.9	86.1

Table 4: Impact of loss components during SFDA-SPF on GlaS

\rightarrow

C17-0 with PixelCAM.

Panel of Figure 5 — Figure 5: Ablation analysis of SFDA-SPF on GlaS $\rightarrow$ C17-3 with PixelCAM . (a) Impact of the localization loss during adaptation. (b) Comparison between static sets , where forget and retain sets are constructed once before adaptation, and dynamic resampling are periodically reconstructed.

Figure 6: Evolution of retain- and forget-set selection accuracy during GlaS

\rightarrow

C17-4 adaptation with DeepMIL . The forget and retain sets are dynamically reconstructed every five epochs.

Appendix figures & tables35 assets

Supplementary material from the paper’s appendix.

Appendix

	GlaS	C16	C17-0	C17-1	C17-2	C17-3	C17-4
Train	67	24348	634	1066	498	816	940
Val (CL)	18	8850	144	38	110	102	146
Val ( PxAP )	6	10	10	10	10	10	10
Test	80	15664	262	172	448	376	298

Appendix

Table 5: Dataset splits used in our experiments, showing the number of patches per split for each dataset and center.

Figure 8: Representative cancer examples from the training split of the constructed CAMELYON17 WSOL benchmark across the five centers. Ground-truth tumor regions are overlaid in red.

Figure 9: Representative cancer examples from the testing split of the constructed CAMELYON17 WSOL benchmark across the five centers. Ground-truth tumor regions are overlaid in red.

Class	Dataset	Source/Target	Train	Val	Test
Normal	152	76	46	15	15
Polyp	948	474	284	95	95
Low-grade IN	1,276	638	382	128	128
High-grade IN	372	186	112	37	37
Serrated adenoma	116	58	34	12	12
Adenocarcinoma	1,590	795	477	159	159

Appendix

Table 6: Class distribution of the EBHI dataset and the constructed source/target splits.

Category	Corruption	Parameters
Staining	Stain-Light	$\theta=0.05$
	Stain-Heavy	$\theta=0.10$
Contamination	Dust	intensity $=0.7$ , $N_{\mathrm{smudges}}\in[3,10]$
	Air Bubble	$N_{\mathrm{bubbles}}\in[10,20]$ , $\alpha=0.3$ , blur severity $=2$
Blurring	Defocus Blur	severity $=5$ , radius $=10$ , alias blur $=0.5$
	Motion Blur	severity $=5$ , radius $=20$ , $\sigma=15$

Appendix

Table 7: Parameters used to construct the corrupted EBHI-Seg target domains. We follow the official Histopath-C implementation [ Noori et al., 2026 ] .

Figure 10: Examples from the training split of the constructed EBHI-S WSOL benchmark.

Figure 11: Examples from the testing split of the constructed EBHI-S WSOL benchmark.

Figure 12: Examples from the testing split of the constructed CAMELYON17 WSOL benchmark.

Hyperparameter	Value
Backbones	ResNet50 / DeiT
Optimizer	SGD
Nesterov acceleration	True
Momentum	${\in\{0.1,0.4,0.9\}}$ for source models
Weight decay	$0.0001$
Learning rate	${\in\{0.0001,0.001,0.002,0.01,0.02,0.1\}}$ for source

Appendix

Table 8: General hyperparameters used during source model training.

Hyperparameter	Value
Momentum	0.9
Weight decay	$0.0001$
Learning rate	${\in\{0.000001,0.00001,0.0001,0.001\}}$
Learning rate decay	CAMELYON17 : 0.1 ${\in\{20\}}$ epochs
Learning rate decay	CAMELYON16 : 0.1 ${\in 5}$ epochs
Flip ratio (forgotten samples)	${\in\{0.05,0.15\}}$

Appendix

Table 9: General hyperparameters used during SFDA training.

Hyperparameter	Value
SHOT	${\lambda_{ent}\in\{0.5,1.0\}}$
	${\lambda_{div}\in\{0.5,1.0\}}$
	${\lambda_{peu}\in\{0.5,1.0\}}$
CDCL	${\tau\in\{0.04,0.08,0.5\}}$
CDCL	${\lambda_{cdcl}\in\{0.5,1.0\}}$
SFDE	${\tau\in\{0.8,1.0\}}$

Appendix

Table 10: Per-method hyperparameters for SFDA methods.

Method	C17-0 $\rightarrow$ GlaS	C17-1 $\rightarrow$ GlaS	C17-2 $\rightarrow$ GlaS	C17-3 $\rightarrow$ GlaS	C17-4 $\rightarrow$ GlaS	Average
SHOT	52.50 / 62.14	55.00 / 59.33	48.75 / 56.84	53.75 / 66.37	53.75 / 56.83	52.75 / 60.30
SFDA-DE	53.75 / 63.54	50.00 / 56.22	55.00 / 55.90	53.75 / 66.37	38.75 / 59.71	50.25 / 60.35
CDCL	53.75 / 63.64	46.25 / 52.55	47.50 / 55.60	53.75 / 64.25	46.25 / 52.70	49.50 / 57.75
ERL	53.75 / 56.92	70.00 / 59.40	56.25 / 56.21	53.75 / 43.52	51.25 / 60.24	57.00 / 55.26
RGV	53.75 / 64.84	53.75 / 50.40	55.00 / 55.90	53.75 / 63.72	53.75 / 60.62	54.00 / 59.10
SFDA-SPF	53.75 / 53.11	77.50 / 56.62	61.25 / 59.83	53.75 / 62.04	55.00 / 59.87	60.25 / 58.29

Appendix

Table 11: Reverse cross-dataset adaptation from each CAMELYON17 center to GlaS . Each entry reports classification accuracy/localization PxAP . The last column reports the average across the five CAMELYON17 source centers.

Method	$50{:}50$	$60{:}40$	$70{:}30$	$80{:}20$	$90{:}10$	Average
	CL / PxAP	CL / PxAP	CL / PxAP	CL / PxAP	CL / PxAP	CL / PxAP
CDCL	50.0 / 25.3	44.6 / 23.4	48.1 / 25.3	46.6 / 25.2	46.6 / 25.5	47.2 / 24.9
SHOT	70.0 / 19.2	41.2 / 37.2	41.2 / 37.2	41.2 / 37.2	41.2 / 37.2	47.0 / 33.6
ERL	59.5 / 13.2	43.1 / 28.2	44.3 / 21.9	43.9 / 22.1	43.9 / 21.7	46.9 / 21.4
RGV	50.0 / 31.0	50.0 / 34.2	50.0 / 34.3	50.0 / 34.9	50.0 / 35.3	50.0 / 33.9
SFDA-SPF	86.2 / 49.9	82.1 / 47.6	83.2 / 50.9	83.9 / 49.7	84.7 / 50.2	84.0 / 49.7

Appendix

Table 12: Robustness to cancer-majority class imbalance in the target adaptation (train) set on CAMELYON17 . Starting from the balanced setting (

50{:}50

), we progressively increase the Cancer/Normal ratio up to

90{:}10

. Results are reported as classification accuracy ( CL ) / localization PxAP (%). Best results are shown in bold .

Target	$p=0.01$	$p=0.05$	$p=0.10$
	CL / PxAP	CL / PxAP	CL / PxAP
C17-0 $\rightarrow$ 1	67.44 / 25.37	74.42 / 34.69	75.58 / 40.59
C17-0 $\rightarrow$ 2	78.35 / 48.17	79.24 / 54.34	79.02 / 59.12
C17-0 $\rightarrow$ 3	92.29 / 63.45	93.88 / 60.44	93.09 / 62.39
C17-0 $\rightarrow$ 4	83.22 / 46.13	79.87 / 48.88	77.85 / 47.43

Appendix

Table 13: Sensitivity to the forgetting proportion

p

under inter-center adaptation on CAMELYON17 . The source model is trained on center 0 and independently adapted to centers 1–4. Results are reported as classification accuracy (CL) / localization PxAP (%). Best results for each target are shown in bold .

Target	Ratio of Images for Localization
	$0.2$	$0.5$	$1.0$
	CL / PxAP	CL / PxAP	CL / PxAP
C17-1	68.02 / 37.95	75.58 / 40.59	74.42 / 34.69
C17-2	79.24 / 59.38	79.02 / 53.47	79.24 / 54.34
C17-3	92.29 / 63.45	93.35 / 60.61	93.88 / 60.44
C17-4	83.22 / 46.13	78.52 / 38.03	79.87 / 48.88

Appendix

Table 14: Sensitivity to the localization keep ratio on CAMELYON17 inter-center adaptation using center 0 as source. Results are reported as classification accuracy ( CL ) / localization PxAP (%).

Backbone	Classifier	PxAP $\uparrow$	CL $\uparrow$
Frozen	Trainable	31.4	47.7
Trainable	Frozen	31.4	85.8
Trainable	Trainable	49.9	86.1

Appendix

Table 15: Impact of jointly adapting the backbone and image-level classifier. Results are reported on GlaS

\rightarrow

C17-0 with PixelCAM . The source model used for CAM-based pixel pseudo-label generation remains frozen in all configurations.

Figure 13: Entropy distributions of correctly and incorrectly classified samples predicted as cancer by SAT.

	SHOT (ICML’20)		CDCL (TMM’22)		ERL (ICLR’23)		RGV (CVPR’25)		Ours
Target condition	PxAP	F1	PxAP	F1	PxAP	F1	PxAP	F1	PxAP	F1
Target clean	83.7	77.2	83.7	77.2	83.7	77.2	83.7	77.2	89.3	82.3
Stain-light	83.5	75.1	83.8	67.1	83.7	69.9	83.8	67.1	90.9	75.1
Stain-heavy	86.7	70.3	81.3	55.9	70.5	56.7	80.8	43.9	82.7	57.9
Dust	82.9	79.4	83.1	71.0	83.1	71.0	83.1	71.0	86.7	72.4
Air bubble	77.8	78.1	76.8	45.3	76.8	45.3	76.8	45.3	78.4	66.5

Appendix

Table 16: Localization ( PxAP ) and classification (F1) performance on EBHI-T under controlled domain shifts. The source model is trained on the clean source domain and independently adapted to each target condition. For all methods, results are reported at the checkpoint selected according to the best F1 score. Best results are shown in bold and second-best results are underlined .

			Target Clean (Shift 1)				Brightness (Shift 10)				Defocus Blur (Shift 6)				Contrast (Shift 11)
Class	GT	GT (%)	Rec.	Prec.	F1	Pred.	Rec.	Prec.	F1	Pred.	Rec.	Prec.	F1	Pred.	Rec.	Prec.	F1	Pred.
Normal	22	3.3	13.6	100.0	24.0	3 (0.5%)	18.2	80.0	29.6	5 (0.8%)	0.0	0.0	0.0	0 (0.0%)	0.0	0.0	0.0	0 (0.0%)
Polyp	143	21.4	94.4	83.9	88.8	161 (24.1%)	95.1	81.0	87.5	168 (25.2%)	10.5	65.2	18.1	23 (3.4%)	6.3	81.8	11.7	11 (1.6%)
Low-grade IN	191	28.6	91.1	91.1	91.1	191 (28.6%)	81.2	92.8	86.6	167 (25.0%)	0.0	0.0	0.0	0 (0.0%)	0.0	0.0	0.0	0 (0.0%)
High-grade IN	55	8.2	69.1	73.1	71.0	52 (7.8%)	69.1	64.4	66.7	59 (8.8%)	90.9	9.4	17.0	535 ( 80.2% )	98.2	8.8	16.1	614 ( 92.1% )
Serrated adenoma	17	2.5	94.1	94.1	94.1	17 (2.5%)	94.1	94.1	94.1	17 (2.5%)	5.9	100.0	11.1	1 (0.1%)	0.0	0.0	0.0	0 (0.0%)

Appendix

Table 17: Per-class source-model predictions under representative target shifts on the EBHI training set. For each class, we report the ground-truth distribution (GT), per-class recall (Rec.), precision (Prec.), F1-score, and predicted class distribution (Pred.). Target Clean and Brightness represent conditions where the source model preserves predictions across classes, whereas Defocus Blur and Contrast illustrate severe prediction collapse.

Figure 14: Reliability of high-confidence foreground regions for GlaS $\rightarrow$ CAMELYON17 . For each positive target image, CAM pixels are ranked according to their activation score, and foreground precision is computed as the fraction of top-ranked pixels overlapping the ground-truth foreground mask. Results are reported for PixelCAM, SAT, and DeepMIL across the five CAMELYON17 centers.

Figure 15: Reliability of high-confidence foreground regions under CAMELYON17 inter-center domain shifts using center 0 as the source. The source models are evaluated on centers 1–4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 16: Reliability of high-confidence foreground regions under CAMELYON17 inter-center domain shifts using center 1 as the source. The source models are evaluated on centers 0, 2, 3, and 4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 17: Reliability of high-confidence foreground regions under CAMELYON17 inter-center domain shifts using center 2 as the source. The source models are evaluated on centers 0, 1, 3, and 4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 18: Reliability of high-confidence foreground regions under CAMELYON17 inter-center domain shifts using center 3 as the source. The source models are evaluated on centers 0, 1, 2, and 4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 19: Reliability of high-confidence foreground regions under CAMELYON17 inter-center domain shifts using center 4 as the source. The source models are evaluated on centers 0–3. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 20: Reliability of low-activation background regions for GlaS $\rightarrow$ CAMELYON17 . For each positive target image, CAM pixels are ranked according to their activation score, and background precision is computed as the fraction of the lowest-activation pixels overlapping the ground-truth background region. Results are reported for PixelCAM, SAT, and DeepMIL across the five CAMELYON17 centers.

Figure 21: Reliability of low-activation background regions under CAMELYON17 inter-center domain shifts using center 0 as the source. The source models are evaluated on centers 1–4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 22: Reliability of low-activation background regions under CAMELYON17 inter-center domain shifts using center 1 as the source. The source models are evaluated on centers 0, 2, 3, and 4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 23: Reliability of low-activation background regions under CAMELYON17 inter-center domain shifts using center 2 as the source. The source models are evaluated on centers 0, 1, 3, and 4. Results are reported for PixelCAM, SAT, and DeepMIL.

Figure 24: Distribution of correct and incorrect predictions across predictive-entropy bins under different domain shifts and target class priors. Results are shown for PixelCAM . Bars report the percentage of target samples within each normalized entropy bin, decomposed into correctly (blue) and incorrectly (orange) classified samples.

Figure 25: Relationship between prediction entropy and classification accuracy under CAMELYON17 inter-center domain shifts using center 2 as the source. Results are reported for PixelCAM .

Figure 26: Relationship between prediction entropy and classification accuracy under CAMELYON17 inter-center domain shifts using center 2 as the source. Results are reported for SAT .

Figure 27: Localization produced by the PixelCAM WSOL model after SFDA-SPF and the state-of-the-art SFDA methods ( GlaS $\rightarrow$ CAMELYON17-3 ).

Figure 28: Localization produced by the SAT WSOL model after SFDA-SPF and the state-of-the-art SFDA methods ( GlaS $\rightarrow$ CAMELYON17-3 ).

Figure 29: Localization produced by the DeepMIL WSOL model after SFDA-SPF and the state-of-the-art SFDA methods ( GlaS $\rightarrow$ CAMELYON17-3 ).

Source-Free Domain Adaptation

Source-Free Domain Adaptation transfers models to new data distributions without access to the original training examples.

See all 69 papers

AdaptMIL: Domain Adaptation for Multiple Instance Learning in Digital Pathology
Oct 20, 2026Ben Vardi, Dana Schonberger, Yuval Friedmann +4Domain Adaptation Computational Pathology
Reichman University · Institute of Pathology, Sheba Medical Center · Technion – Israel Institute of Technology +1
Discrepancy Minimization Improves Cross-Hospital Robustness in Digital Pathology
May 24, 2026Ben Vardi, Dana Schonberger, Yuval Friedmann +4Computational Pathology Unsupervised Domain Adaptation
Reichman University, Herzliya, Israel · Institute of Pathology, Sheba Medical Center, Ramat-Gan, Israel · Technion - Israel Institute of Technology, Haifa, Israel
Single-Stage Hierarchical Rectification for Weakly Supervised Histopathology Segmentation
Jun 18, 2026Duc T. Nguyen, Hoang-Long Nguyen, Thanh-Ha DO +1Computational Pathology Weakly Supervised Image Segmentation
College of Engineering & Computer Science, VinUniversity, Hanoi, Vietnam · VinUni-Illinois Smart Health Center, VinUniversity, Hanoi, Vietnam · The Computer Vision and Medical AI Lab, VinUniversity, Hanoi, Vietnam +1

Oct 20, 2026J/K move · Enter open · S save

AdaptMIL: Domain Adaptation for Multiple Instance Learning in Digital Pathology

Ben Vardi, Dana Schonberger, Yuval Friedmann, Zohar Yakhini+3

Reichman University · Institute of Pathology, Sheba Medical Center · Technion – Israel Institute of Technology+1

Domain Adaptation Computational Pathology Multiple Instance Learning Pathology Foundation Models

Open paper AskPDF arXiv

Pathology foundation models (PFMs) support a wide range of histopathology tasks on whole-slide images through multiple instance learning (MIL). Despite recent progress, these models remain sensitive to cross-hospital distribution shifts: performance often degrades when a task head (e.g., classification) is trained on data from one hospital and then used on data from another hospital. To address this challenge, we propose AdaptMIL, a training strategy for integrating domain adaptation into MIL training. AdaptMIL is motivated by our finding that directly applying domain adaptation objectives during MIL training can interfere with MIL aggregation learning, often resulting in performance degradation over standard training. AdaptMIL resolves this by decoupling representation-level domain alignment learning from MIL aggregation learning through a two-stage training procedure. We show that AdaptMIL improves cross-hospital performance and outperforms alternative adaptation methods across experiments spanning multiple WSI datasets, tasks, PFMs, MIL architectures, and domain adaptation objectives. AdaptMIL's code will be released upon publication.

Abstract

Source-Free Domain Adaptation

Explore similar work

AdaptMIL: Domain Adaptation for Multiple Instance Learning in Digital Pathology

Discrepancy Minimization Improves Cross-Hospital Robustness in Digital Pathology

Single-Stage Hierarchical Rectification for Weakly Supervised Histopathology Segmentation