# Decoupled Pipeline with Proposal Reranking and Score Fusion for Positive-Unlabeled Marine Species Detection

arXiv: https://arxiv.org/abs/2607.18700 · HTML: https://mlsanity.com/paper/2607.18700/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 19 references that are in the map.
- First posted on arXiv 21 Jul 2026.

## Builds on (most cited first, 8 of 8)

- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection](https://mlsanity.com/paper/2202.06934/) (2022)
- [BioCLIP 2: Emergent Properties from Scaling Hierarchical Contrastive Learning](https://mlsanity.com/paper/2505.23883/) (2025)
- [FathomNet: A global image database for enabling artificial intelligence in the ocean](https://mlsanity.com/paper/2109.14646/) (2021)
- [Tile-Based ViT Inference with Visual-Cluster Priors for Zero-Shot Multi-Species Plant Identification](https://mlsanity.com/paper/2507.06093/) (2025)
- [DS@GT AnimalCLEF: Triplet Learning over ViT Manifolds with Nearest Neighbor Classification for Animal Re-identification](https://mlsanity.com/paper/2509.12353/) (2025)
- [The Label Imitation Game: Turing Test Network for Zero-Shot Pseudo-Label Pruning](https://mlsanity.com/paper/2606.30875/) (2026)

## Abstract (arXiv)

The FathomNetCLEF 2026 competition combines underwater object detection and fine-grained marine species classification under a positive-unlabeled evaluation setting. The provided training labels are sparse, while the hidden test set is out-of-distribution relative to the training imagery, creating both annotation incompleteness and source-shift challenges. We describe DS@GT ARC's multi-stage system developed for this setting while keeping model training restricted to the data provided by the competition. The final private-leaderboard model uses a frozen Megalodon YOLOv8x detector as a class-agnostic proposal generator, combines global and tiled inference with tile-edge filtering, classifies expanded proposal crops with a LoRA-finetuned DINOv3 ViT-H classifier, and ranks predictions using weighted geometric fusion of detector and classifier confidence. This system placed 12th out of 102 teams. A closely related variant added a locally trained TTN-inspired validity head as a light reranking signal, improving public-leaderboard and proxy-evaluation performance but slightly reducing private-leaderboard performance. Across experiments, the strongest lesson was that train-derived validation and detector-only metrics were not reliable enough for model selection. Instead, we used proxy datasets only for validation and comparison, and combined those signals with leaderboard feedback and targeted ablations. These experiments showed that reserving proposal recall, avoiding over-aggressive filtering, and improving downstream ranking were more effective than fine-tuning the detector or directly training on noisy pseudo-labels. Code: https://github.com/dsgt-arc/fathomnetclef-2026.