- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 3
Read review
Paper summary
The paper proposes AeroEVA, a Qwen3-VL-4B agent for active geo-localization: navigating aerial grids with limited moves and local observations to reach image- or text-specified goals. Its Envision–Verify–Act framework predicts upcoming views, scores candidate moves by similarity to the goal image, and uses these scores to guide actions. Training combines joint SFT on shortest paths with multi-turn GRPO. On MASA, AeroEVA achieves 71.8% success with SFT and 86.1% with GRPO, outperforming retrained GOMAA-Geo (65.0%) and same-backbone GeoExplorer (47.9%). The checkpoint transfers to unseen regions, post-disaster imagery, larger grids, and ground-photo and text goals. The paper also introduces AeroTask, comprising 696 LLM-generated task-style language goals.
Critical strengths
- Well-supported design choices. Diagnostics support joint supervision and nearest-row prediction, while ablations show that visual prediction helps only when paired with text targets and verification.
- Strong gains and broad transfer. AeroEVA outperforms same-backbone controls (77.0 vs. 56.4) and generalizes across regions, time, goal types, and grid sizes, reaching 60.9% on 10×10 grids versus at most 12.9% for baselines.
Critical weaknesses
- The "unified" part is not isolated. The verifier is a separate frozen snapshot, so it could exist without visual prediction in the policy itself. Table 2 has no "Action SFT + text + tool" row without visual-latent supervision, which is needed to show that the policy's own prediction matters.
- The verifier's share of the gain is unknown. I could not find MASA results for the final model with the tool turned off on aerial-image goals. Tool-off results on other goal types are encouraging but not a controlled comparison.
- No training-seed variance. All results come from one training run; the reported ± values reflect only decoding randomness. Multi-turn GRPO is often sensitive to seeds.
- Training uses privileged supervision. AeroEVA learns from shortest-path action labels and text targets built from offline land-cover maps, which the RL baselines do not use. The paper should state which gains depend on the map-derived labels.
- AeroTask is weakly validated. The same model (Qwen3-VL-32B) generates and judges the tasks on segmentation pseudo-labels, with no human check of whether each instruction uniquely identifies the target cell. With 268 evaluation records, label noise could affect the conclusions.
Questions for authors
- Can you add an "Action SFT + text targets + frozen verifier" variant without visual-latent loss on the policy?
- How does the final SFT+GRPO model perform on MASA aerial-image goals with the tool turned off?
- Can you report at least 3 training seeds for AeroEVA and the main same-backbone baselines?
- How does a variant perform whose text targets come from the image alone, without land-cover maps?
- Was any part of AeroTask checked by humans for target uniqueness, and what was the agreement rate?