Skip to main content

Indian Institute of Science Education and Research, Bhopal

2 related articles

Subscribe to updates

arXiv Oct 5, 2026
FairRSFM evaluates three remote sensing foundation models by biome group and finds aggregate metrics mask ecological performance gaps
The authors introduce FairRSFM, which maps georeferenced samples from four remote sensing downstream datasets (m-EuroSAT, m-BigEarthNet, m-SA-Crop-Type, and MMEarth20K) to 14 terrestrial biome classes consolidated into six macro-groups, and evaluates Prithvi-EO-2.0, SatMAE, and DOFA under a unified frozen-backbone protocol across three random seeds, showing that aggregate metrics consistently mask biome-dependent disparities, for example Prithvi-EO-2.0 reaching high overall macro-F1 on m-EuroSAT while its worst-group score is markedly lower, and compares BOLP, DBR, and GroupDRO as mitigation baselines whose effectiveness is model- and task-dependent.
The authors introduce FairRSFM, which maps georeferenced samples from four remote sensing downstream datasets (m-EuroSAT, m-BigEarthNet, m-SA-Crop-Type, and MMEarth20K) to 14 terrestrial biome classes consolidated into six macro-groups, and evaluates Prithvi-EO-2.0, SatMAE, and DOFA under a unified frozen-backbone protocol across three random seeds, showing that aggregate metrics consistently mask biome-dependent disparities, for example Prithvi-EO-2.0 reaching high overall macro-F1 on m-EuroSAT while its worst-group score is markedly lower, and compares BOLP, DBR, and GroupDRO as mitigation baselines whose effectiveness is model- and task-dependent.
Indian Institute of Technology Bombay Indian Institute of Science Education and Research, Bhopal
Md Aminur Hossain（First author）
Linear Probing Distributionally Robust Optimization Benchmark Construction and Evaluation Foundation Models
arXiv Sep 24, 2026
Compact domain footprints in a frozen embedding space enable generative replay for continual pathology report generation without storing slides or patch exemplars, outperforming exemplar-free and limited-buffer rehearsal baselines on multiple public continual learning benchmarks
The work introduces an exemplar-free continual learning framework for whole-slide-image-to-report generation: it builds a compact domain footprint per domain in a frozen patch-embedding space (a k-means codebook, a slide-level code histogram bank, patch-count statistics, and a report-style prototype), uses it to synthesize pseudo-WSIs whose pseudo-reports come from an immediate teacher snapshot for generative replay, and conditions the language model through a style prefix; across multiple public continual learning benchmarks the approach outperforms exemplar-free and limited-buffer rehearsal baselines and supports domain-agnostic inference without explicit domain identifiers.
The work introduces an exemplar-free continual learning framework for whole-slide-image-to-report generation: it builds a compact domain footprint per domain in a frozen patch-embedding space (a k-means codebook, a slide-level code histogram bank, patch-count statistics, and a report-style prototype), uses it to synthesize pseudo-WSIs whose pseudo-reports come from an immediate teacher snapshot for generative replay, and conditions the language model through a style prefix; across multiple public continual learning benchmarks the approach outperforms exemplar-free and limited-buffer rehearsal baselines and supports domain-agnostic inference without explicit domain identifiers.
University of Regensburg Indian Institute of Information Technology Bhopal Indian Institute of Science Education and Research, Bhopal
Pratibha Kumari（First author, Corresponding author）
Histopathology Report Generation Domain Footprint Generative Replay Continual Learning Report-Style Prototype

You’re all caught up