# A Parameter-Efficient Mixture-of-Experts Framework for Cross-Modal Geo-Localization

arXiv: https://arxiv.org/abs/2510.20291 · HTML: https://mlsanity.com/paper/2510.20291/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts.
- Builds on 21 of its 38 references that are in the map.
- First posted on arXiv 23 Oct 2025.

## Placed in

- Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/)

## Builds on (most cited first, 21 of 21)

- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts](https://mlsanity.com/paper/2111.02358/) (2021)
- [Long-CLIP: Unlocking the Long-Text Capability of CLIP](https://mlsanity.com/paper/2403.15378/) (2024)
- [Alpha-CLIP: A CLIP Model Focusing on Wherever You Want](https://mlsanity.com/paper/2312.03818/) (2023)
- [EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters](https://mlsanity.com/paper/2402.04252/) (2024)
- [TF-CLIP: Learning Text-free CLIP for Video-based Person Re-Identification](https://mlsanity.com/paper/2312.09627/) (2023)
- [Where am I? Cross-View Geo-localization with Natural Language Descriptions](https://mlsanity.com/paper/2412.17007/) (2024)
- [Visual-Text Cross Alignment: Refining the Similarity Score in Vision-Language Models](https://mlsanity.com/paper/2406.02915/) (2024)
- [Re-Align: Aligning Vision Language Models via Retrieval-Augmented Direct Preference Optimization](https://mlsanity.com/paper/2502.13146/) (2025)
- [CLIP-CID: Efficient CLIP Distillation via Cluster-Instance Discrimination](https://mlsanity.com/paper/2408.09441/) (2024)
- [Cross-Modal and Uni-Modal Soft-Label Alignment for Image-Text Retrieval](https://mlsanity.com/paper/2403.05261/) (2024)
- [Vision-Language Models for Autonomous Driving: CLIP-Based Dynamic Scene Understanding](https://mlsanity.com/paper/2501.05566/) (2025)
- [WeatherPrompt: Multi-modality Representation Learning for All-Weather Drone Visual Geo-Localization](https://mlsanity.com/paper/2508.09560/) (2025)
- [BLIP: Facilitating the Exploration of Undesirable Consequences of Digital Technologies](https://mlsanity.com/paper/2405.06783/) (2024)
- [EvdCLIP: Improving Vision-Language Retrieval with Entity Visual Descriptions from Large Language Models](https://mlsanity.com/paper/2505.18594/) (2025)
- [DiscoVLA: Discrepancy Reduction in Vision, Language, and Alignment for Parameter-Efficient Video-Text Retrieval](https://mlsanity.com/paper/2506.08887/) (2025)
- [RA-BLIP: Multimodal Adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training](https://mlsanity.com/paper/2410.14154/) (2024)
- [Astrea: A MOE-based Visual Understanding Model with Progressive Alignment](https://mlsanity.com/paper/2503.09445/) (2025)
- [LowCLIP: Adapting the CLIP Model Architecture for Low-Resource Languages in Multimodal Image Retrieval Task](https://mlsanity.com/paper/2408.13909/) (2024)

## Built on by (most cited first, 1 of 1)

- [Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization](https://mlsanity.com/paper/2609.18139/) (2026)

## Abstract (arXiv)

We present a winning solution to RoboSense 2025 Track 4: Cross-Modal Drone Navigation. The task retrieves the most relevant geo-referenced image from a large multi-platform corpus (satellite/drone/ground) given a natural-language query. Two obstacles are severe inter-platform heterogeneity and a domain gap between generic training descriptions and platform-specific test queries. We mitigate these with a domain-aligned preprocessing pipeline and a Mixture-of-Experts (MoE) framework: (i) platform-wise partitioning, satellite augmentation, and removal of orientation words; (ii) an LLM-based caption refinement pipeline to align textual semantics with the distinct visual characteristics of each platform. Using BGE-M3 (text) and EVA-CLIP (image), we train three platform experts using a progressive two-stage, hard-negative mining strategy to enhance discriminative power, and fuse their scores at inference. The system tops the official leaderboard, demonstrating robust cross-modal geo-localization under heterogeneous viewpoints.