This checkpoint is designed for SwiftVLN on SatNav, the continuous-state vision-and-language navigation benchmark over satellite imagery. It augments visual tokens with relative pose through feature-wise linear modulation.

Model

Starting checkpoint: Qwen/Qwen2.5-VL-3B-Instruct
Training data: SatNav-v0.1 offline expert trajectories
Training: 1 epoch, full-parameter fine-tuning, learning rate 2e-5
Context: 32 RGB frames and up to 8 uniformly sampled history frames
Prediction horizon: 4 actions
Memory: per-frame average pooling with stride 2
Pose fusion: FiLM over relative position and heading features

Repository name

swiftvln-satnav: SwiftVLN trained and evaluated on SatNav
3b: Qwen2.5-VL 3B backbone
1ep: trained for one epoch
f32s4: uses a 32-frame window and predicts four actions
overlap0: uses non-overlapping training windows
pf-h8-pool-s2: uses the reference per-frame memory configuration
posefilm: injects relative pose into visual tokens with FiLM

Reference Results

Evaluated in SatSim on the full splits with a 500-step cap.

Split	Episodes	NE	OS	SR	SPL
`val_seen`	4,574	36.59	76.26	67.77	67.50
`val_unseen`	8,756	60.39	65.02	55.42	55.17

Usage

Use this checkpoint with the SwiftVLN evaluation guide. Keep the full repository name unchanged because SwiftVLN derives the evaluation configuration from it.

This checkpoint is an input-augmentation ablation in the SwiftVLN SatNav Ablation Model Zoo.

Eku127/swiftvln-satnav-3b-1ep-f32s4-overlap0-pf-h8-pool-s2-posefilm

SwiftVLN for SatNav — Relative Pose

Related Repositories

Model

Repository name

Reference Results

Usage

Eku127/
swiftvln-satnav-3b-1ep-f32s4-overlap0-pf-h8-pool-s2-posefilm