# Bridging Multimodal Fusion and Expert Routing via Spectral Reliability Descriptors for Robust Object Detection

arXiv: https://arxiv.org/abs/2606.01173 · HTML: https://mlsanity.com/paper/2606.01173/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts.
- Builds on 18 of its 35 references that are in the map.
- First posted on arXiv 31 May 2026.

## Placed in

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

## Builds on (most cited first, 18 of 18)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [Target-aware Dual Adversarial Learning and a Multi-scenario Multi-Modality Benchmark to Fuse Infrared and Visible for Object Detection](https://mlsanity.com/paper/2203.16220/) (2022)
- [Global Filter Networks for Image Classification](https://mlsanity.com/paper/2107.00645/) (2021)
- [VarifocalNet: An IoU-aware Dense Object Detector](https://mlsanity.com/paper/2008.13367/) (2020)
- [CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi-Modality Image Fusion](https://mlsanity.com/paper/2211.14461/) (2022)
- [Dynamic Head: Unifying Object Detection Heads with Attentions](https://mlsanity.com/paper/2106.08322/) (2021)
- [Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning](https://mlsanity.com/paper/2003.02437/) (2020)
- [D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement](https://mlsanity.com/paper/2410.13842/) (2024)
- [DEIM: DETR with Improved Matching for Fast Convergence](https://mlsanity.com/paper/2412.04234/) (2024)
- [Learning a Graph Neural Network with Cross Modality Interaction for Image Fusion](https://mlsanity.com/paper/2308.03256/) (2023)
- [Missing Modality Prediction for Unpaired Multimodal Learning via Joint Embedding of Unimodal Models](https://mlsanity.com/paper/2407.12616/) (2024)
- [Removal then Selection: A Coarse-to-Fine Fusion Perspective for RGB-Infrared Object Detection](https://mlsanity.com/paper/2401.10731/) (2024)
- [TSJNet: A Multi-modality Target and Semantic Awareness Joint-driven Image Fusion Network](https://mlsanity.com/paper/2402.01212/) (2024)
- [FreDFT: Frequency Domain Fusion Transformer for Visible-Infrared Object Detection](https://mlsanity.com/paper/2511.10046/) (2025)
- [AlignFreeNet: Is Cross-Modal Pre-Alignment Necessary? An End-to-End Alignment-Free Lightweight Network for Visible-Infrared Object Detection](https://mlsanity.com/paper/2507.20146/) (2025)

## Abstract (arXiv)

RGB-infrared detectors typically discard the statistics generated during cross-modal fusion, leaving downstream modules unaware of whether the current interaction is reliable. We propose to extract a parameter-free, 7-dimensional spectral reliability descriptor -- summarizing band energy, amplitude ratio, phase consistency, and cross-modal correlation -- and to reuse it beyond the fusion stage. The descriptor drives both Spectral Reliability Fusion (SRF), which gates a spectral residual against a conservative spatial base, and Reliability-Conditioned Expert Routing (RCER), which combines the descriptor with pooled content to steer sparse post-fusion experts. Under matched ablations, descriptor-aware gating improves mAP50 over content-only adaptive gating; a 2×2 factorial analysis further shows that descriptor-conditioned routing provides the larger marginal gain over expert architecture alone at near-equal parameter count. Under six synthetic degradations on DroneVehicle, average retention rises to 95.0%, versus 92.0% for content-only MoE and 87.9% for concatenation, with the largest gain under modality drop; the same model also improves mAP50 by +5.2/+5.3 on the natural day/night split. These results suggest that preserving fusion-time reliability as an explicit signal benefits both adaptive fusion and post-fusion conditional computation.