Pith. sign in

REVIEW 4 major objections 6 minor 57 references

AirForesight: Current-to-Future Spatial Map Imagination with Cross-Space Planning Consistency for UAV-VLN

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read AirForesight shows that a UAV navigates more successfully when, before each waypoint, it forms a latent map of its current scene, imagines a future version of that map, and is trained so the imagined map-space direction agrees with the…

desk verdict Solid, well-ablated UAV-VLN paper with a real but modest gain; the main blemish is tuning on the test split, not the pseudo-label pipeline. read the letter →

arxiv 2608.12835 v1 pith:X7AQDDC6 submitted 2026-08-13 cs.RO

Yutong Liu , Xiaojie Li , Mingzhu Xu , Jianlong Wu This is my paper · ORCID
classification cs.RO
keywords UAVvision-languagenavigationspatialmapimaginationfuture-mappredictioncross-spaceplanningconsistencylatentrepresentationOpenbenchmarkAerialVLN-Smultimodallargelanguagemodel
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that an aerial vision–language navigation agent navigates more successfully when, before committing to each waypoint, it forms a latent map of its current surroundings, imagines how that map will look a few steps ahead, and is trained so the imagined map-space direction agrees with the expert's flight direction. The authors argue that existing methods either react to observations directly or rely on pre-computed spatial priors, neither of which forces the model to reason about future map states in a way that is grounded in executable motion. On OpenUAV's Test Seen split, AirForesight lifts success rate from 30.47% to 35.83% and success-weighted path length from 25.37% to 30.22% compared with the TravelUAV baseline, with smaller gains on unseen maps; it also reports the highest success rate among compared methods on AerialVLN-S. If the claim holds, future-aware latent map imagination offers a cheap way to inject spatial reasoning into navigation policies without dense online map construction.

What carries the argument

The load-bearing object is the trio of learnable query tokens—⟨cur_map⟩, ⟨fut_map⟩, ⟨waypoint⟩—inserted into a frozen Vicuna-7B and processed under structured causal attention: current-map tokens see only instruction and images, future-map tokens also see current-map tokens, and the waypoint token sees all preceding tokens. A lightweight MAE-style decoder turns the latent map representations into dense features, and shared map heads reconstruct current and future semantic maps while a trajectory head predicts the future trajectory mask on the current map plane. The second mechanism is the cross-space planning consistency loss (CPCL): PCA over nearby predicted-trajectory points yields a map-space direction, which is cosine-aligned to the expert action direction obtained by projecting the ground-truth waypoint displacement onto the same map plane. Together these force the current-map representation to encode both present layout and future motion intent, propagate that knowledge causally into future-map prediction, and anchor map-space reasoning to executable action direction.

What would settle it

Take AirForesight, retrain it on OpenUAV with the spatial supervision intentionally corrupted—for example, shuffle the semantic category labels, replace the future-map target with the current map, or add noise to the depth projection—and compare Test Seen SR. If SR stays near the full model's 35.83%, spatial imagination is not the cause of the gain; if it falls back toward the 31.94% navigation-only baseline, the map supervision is doing the work.

Watch

Extended reading notes

Core claim

AirForesight's central claim is that current-to-future spatial map imagination, learned as latent supervision, improves UAV-VLN. Multi-view RGB and the instruction are tokenized into a Vicuna-7B sequence together with learnable current-map, future-map, and waypoint tokens. Under structured causal attention, current-map tokens encode the present scene and are supervised by both current semantic-map reconstruction and prediction of an n-step future trajectory in map space; future-map tokens then attend to the current-map tokens to predict the map at step t+n; and the waypoint token aggregates everything to predict the next 3D waypoint. A cross-space planning consistency loss takes the dominant PCA direction of the predicted trajectory mask and aligns it, by cosine distance, with the projection of the ground-truth waypoint displacement onto the map plane. The spatial supervision maps come from an offline pseudo-label pipeline, so no dense map is needed at inference. The paper reports best success among compared methods on OpenUAV Test Seen and AerialVLN-S, with ablations attributing the gains to the spatial-map modeling, future imagination, and consistency loss.

Load-bearing premise

The load-bearing premise is that the offline pseudo-label pipeline produces semantic and spatial maps accurate enough to teach navigation: if category extraction or depth-to-grid projection is systematically wrong, the spatial-supervision heads could be learning label artifacts, and the reported gains could shrink or vanish.

Editorial extensions

If this is right

  • On OpenUAV Test Seen Full, AirForesight raises SR from 30.47% to 35.83%, SPL from 25.37% to 30.22%, and cuts NE from 68.44 m to 56.99 m relative to TravelUAV; it also improves every metric on Test Unseen Map (SR 13.67% vs 9.92%).
  • On AerialVLN-S, AirForesight reports SR 14.4 on Validation Seen and 12.4 on Validation Unseen, the highest among compared methods, while keeping the lowest NE on Validation Seen.
  • The ablations show each component contributes: current spatial map modeling alone lifts SR from 31.94% to 35.33%, future spatial imagination adds further gains, and adding CPCL gives the best result while raising cosine similarity between predicted map-space trajectory and expert action direction from 0.903 to 0.960.
  • Because the maps are learned as latent supervision, inference needs no online dense reconstruction: AirForesight runs at 211.51 ms per step versus 1076.39 ms for an online spatial-prior baseline and only 12.31 ms above TravelUAV.
  • Perturbing the learned map tokens at inference drops SR on Test Unseen Map from 13.67% to about 11.4%, indicating the map representations carry navigation-relevant information beyond the base vision-language cues.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the map supervision is offline-only, the framework suggests a general recipe: a navigation policy can be given spatial foresight during training without changing its deployed inputs; the same latent-imagination trick could transfer to ground robots or manipulation policies that currently map pixels to actions.
  • CPCL is a generic directional regularizer between a latent prediction and an expert action; a natural test outside this paper is whether the cosine-alignment loss helps in sharp-turn or sparse-trajectory cases, where the paper itself notes the PCA tangent approximation is least reliable.
  • The reported pseudo-label quality (94.0% category, 82.5% spatial-projection accuracy) implies a testable boundary: varying annotation quality—fewer superclasses, noisier depth, or shuffled categories—should predictably degrade navigation performance if spatial supervision is the active ingredient.
  • A direct extension would condition future-map imagination on the decoded trajectory mask rather than only on the shared current-map representation, converting the current implicit propagation into explicit route-dependent imagination; the paper lists this as future work.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes AirForesight, a framework for UAV vision-and-language navigation that introduces latent current-map and future-map representations into an LLM-based waypoint predictor. Multi-view images and instructions are encoded with Vicuna-7B; learnable current-map, future-map, and waypoint tokens are processed under a structured causal mask. The current-map representation is supervised by reconstructing offline pseudo-label semantic maps and predicting an n-step future trajectory mask; the future-map representation is supervised by reconstructing a future semantic map; the waypoint token then predicts the next 3D waypoint. A cross-space planning consistency loss aligns the PCA direction of the predicted map-space trajectory with the expert action direction derived from the ground-truth waypoint displacement. Training is two-stage, with waypoint prediction first and spatial losses plus CPCL added in the second stage. Experiments on OpenUAV report improved SR/SPL/NE over TravelUAV on Test Seen and gains on Unseen Map/Object splits, with ablations, multi-seed results, latency measurements, and qualitative visualizations; experiments on AerialVLN-S report competitive SR.

Significance. If the reported results hold, the paper makes a useful contribution to UAV-VLN by showing that latent map-space supervision, future-map imagination, and a map-to-action consistency loss can improve waypoint prediction without requiring online dense map construction. The strengths of the paper are the extensive component ablations (Table 3), the multi-seed stability analysis (Table 7), the sensitivity analysis over several design choices (Tables 4, 5, 10, 12), the inference-time perturbation study (Figure 5), and the latency comparison (Table 11). The method is not circular: CPCL is a multi-task consistency constraint between two branches that are both supervised by the same expert displacement, so it does not presuppose the navigation result. The main limitations are that the spatial supervision relies on an imperfect offline pseudo-label pipeline whose error is not tied to navigation performance, and that key hyperparameters are selected on the OpenUAV Test Seen split, which can inflate the headline seen-split numbers. With additional validation experiments and clarifications, the paper would be a solid contribution.

major comments (4)
  1. [Sec. 4.1, Appendix A.3, Table 6]
  2. [Sec. 4.3, Tables 5, 10, 12]
  3. [Sec. 3.4, Eq. (13)]
  4. [Tables 1, 2, and 7]
minor comments (6)
  1. [Throughout] There are several typographical spacing errors in the running headers and table captions, such as 'OpenUA V' and 'UA V-VLN'; please correct these globally.
  2. [Abstract and Sec. 3.4] The phrase 'cross-space planning consistency lossthat' is missing a space before 'that'.
  3. [Sec. 3.3.3, Eq. (7)] The notation using '|' in Eq. (7) and Eq. (9) is nonstandard; please clarify that <cur_map> and <fut_map> denote the output representations of the corresponding token groups rather than token ids.
  4. [Sec. 3.3.1 and Sec. 3.4] Because the map is a fixed-scale grid, please state the spatial extent of the grid in meters and the physical size of each cell, and specify the local neighborhood radius used in the CPCL direction estimation.
  5. [Sec. 4.4.3, Figure 5] For the Gaussian perturbation experiment, please specify the noise magnitude and how the perturbation is applied to the nine map tokens, since the reported SR drop from 13.67% to 11.16% depends on this choice.
  6. [Table 2] Only NE and SR are reported on AerialVLN-S; if OSR and SPL are not available for this benchmark, please state this explicitly so readers do not infer selective reporting.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: AirForesight is a standard supervised multi-task pipeline whose auxiliary losses and CPCL are trained against external ground-truth waypoints and offline-generated pseudo-labels, not against the model's own outputs.

full rationale

I walked each load-bearing step of the derivation chain and found no step that reduces by construction to its own input. The current-map loss (Eq. 4), future-trajectory loss (Eq. 6), and future-map loss (Eq. 8) are supervised by offline pseudo-labels generated by external tools (Gemini, GroundingDINO, MobileSAM) and by projected ground-truth future waypoints. The navigation loss (Eq. 10) is standard behavior cloning on ground-truth waypoint displacement. The cross-space planning consistency loss (Eq. 13) aligns a direction extracted from the predicted trajectory mask with the ground-truth waypoint displacement direction. This is a multi-task consistency objective: both the trajectory head and the waypoint head are independently supervised to the same expert label, so the loss encourages agreement between two supervised branches rather than deriving a prediction from the target it claims to produce. No self-citation is load-bearing: TravelUAV and LLaMA-VID are cited only for implementation components such as the path decoder and visual token compression. No uniqueness theorem or author-imposed ansatz is invoked to force the design. The pseudo-label quality concern is an empirical validity risk about label accuracy, not a circular structure: even imperfect labels still constitute external supervision. I therefore cannot exhibit any specific reduction of Eq. X to Eq. Y by construction, and the paper's central claims remain independent of its inputs.

Assumptions & free parameters 8 free parameters · 4 assumptions · 2 invented entities

The central claim (improved navigation) rests on a set of manually chosen hyperparameters and on the assumption that automatically generated pseudo-label maps are accurate enough to provide spatial supervision. The paper introduces latent map representations as architectural entities without external falsifiable handles; their value is supported only through in-paper ablations and perturbation tests.

free parameters (8)
  • lambda_spatial = 0.5
    Weight balancing spatial losses in stage-2 objective (Eq. 15); chosen by sweeping 0.1, 0.5, 1.0 on OpenUAV Test Seen (Table 5).
  • lambda_cons = 1e-3
    CPCL weight; selected from 1e-2, 1e-3, 1e-4 on Test Seen (Table 10).
  • future_horizon_n = 10
    Number of future steps for trajectory and future map prediction; tuned among 5, 10, 15 on Test Seen (Table 10).
  • map_resolution_HxW = 224x224
    Grid size for semantic map supervision; tuned among 112, 224, 336 on Test Seen (Table 10).
  • num_map_tokens = 9 (per group)
    Number of learnable current/future map query tokens; swept 4, 9, 16 on Test Seen (Table 4).
  • spatial_decoder_depth = 2
    Depth of MAE decoder; swept 1, 2, 4 on Test Seen (Table 12).
  • spatial_decoder_hidden_dim = 768
    Hidden dimension of spatial decoder; swept 512, 768, 1024 on Test Seen (Table 12).
  • class_weights_w1_w2 = not specified numerically
    Weighted cross-entropy weights for map classes and trajectory foreground/background (Eqs. 4, 6); values are not reported, so the balancing scheme is not fully reproducible.
assumptions (4)
  • domain assumption The offline pseudo-label maps derived from Gemini 3 Pro, GroundingDINO, MobileSAM, and depth projection are accurate enough to supervise spatial heads.
    Invoked in Secs. 3.3.1 and A.3; the paper reports approximate category accuracy 94.0% and spatial-projection accuracy 82.5% on 100 manually checked trajectories, but does not quantify the impact of label noise on navigation.
  • domain assumption The PCA-based local direction from the predicted future-trajectory mask is a valid proxy for short-horizon navigation intent.
    Used in Sec. 3.4, Eq. (13); acknowledged in Limitations D as being unreliable for sharply curved paths or sparse masks.
  • domain assumption The expert action direction derived from ground-truth waypoint displacement provides a stable reference for both the waypoint head and CPCL.
    Used in Eqs. (10) and (12); the stability claim is argued but not proven.
  • domain assumption The future semantic map M_{t+n} and future trajectory T_t can be generated from future ground-truth waypoints and future observations in the current UAV-local frame, assuming the scene semantics are consistent over the horizon.
    Described in Secs. 3.3.2 and 3.3.3; no explicit handling of dynamic objects or changes in occlusions.
invented entities (2)
  • Latent current-map representation (Z_cur_map)
    purpose: To encode current scene structure and future motion intent in the LLM token space for waypoint prediction.
    Introduced in Sec. 3.3.1; its role is evidenced only by in-paper ablations and perturbation analysis (Fig. 5), not by an external measurable prediction.
  • Latent future-map representation (Z_fut_map)
    purpose: To anticipate the spatial state at step t+n and integrate it into waypoint prediction.
    Introduced in Sec. 3.3.3; no external handle outside the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of AirForesight: Current-to-Future Spatial Map Imagination with Cross-Space Planning Consistency for UAV-VLN." pith.science (2026). https://pith.science/paper/X7AQDDC6

@misc{pith2026260812835,
  author       = {Pith},
  title        = {Pith review of: AirForesight: Current-to-Future Spatial Map Imagination with Cross-Space Planning Consistency for UAV-VLN},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X7AQDDC6}},
  note         = {Machine review of arXiv:2608.12835}
}
read the original abstract

Unmanned Aerial Vehicle Vision-Language Navigation (UAV-VLN) requires agents to follow language instructions, infer spatial structure from sparse multi-view observations, and execute feasible 3D motion in complex outdoor environments. Despite recent progress with large language models, most existing methods still map vision-language inputs directly to actions, providing limited explicit scene grounding and future-aware spatial reasoning. We propose AirForesight, a current-to-future spatial map imagination framework for UAV-VLN. AirForesight first learns a structured current-map representation from multi-view observations. This representation is jointly supervised by current-map reconstruction and future-trajectory prediction, encouraging it to encode both present scene structure and future motion intent. Under structured causal attention, the current spatial knowledge is propagated to future-map reasoning, and the resulting current and future representations are aggregated to predict the next 3D waypoint. To make spatial imagination more relevant to navigation, we introduce a cross-space planning consistency loss that encourages directional agreement between the predicted map-space trajectory and the expert action direction derived from the ground-truth waypoint displacement. Experiments on OpenUAV and AerialVLN-S, together with extensive ablations, demonstrate strong performance and support the effectiveness and stability of the proposed framework.

Figures

Figures reproduced from arXiv: 2608.12835 by the authors.

Figure 1
Figure 1. Comparison of existing paradigms for UAV-VLN. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overall architecture of AirForesight. The model jointly processes multimodal inputs with learnable current-map, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Visualization of learned spatial representations. (a) Predicted current maps, future trajectories, and future maps, [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison on a challenging long-horizon UAV-VLN episode. The top and bottom rows show representative [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Effect of perturbing the learned map query tokens [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Structured causal attention in AirForesight. Current [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Additional qualitative comparisons across diverse long-horizon UAV-VLN episodes. For each example, the top row [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: Additional current-map, future-trajectory, and future-map predictions with their corresponding supervision targets. [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 35 canonical work pages

  1. [1]

    Dong An, Yuankai Qi, Yangguang Li, Yan Huang, Liang Wang, Tieniu Tan, and Jing Shao. 2022. Bevbert: Multimodal map pre-training for language-guided navigation.arXiv preprint arXiv:2212.04385(2022)

    arXiv 2022

  2. [2]

    Dong An, Hanqing Wang, Wenguan Wang, Zun Wang, Yan Huang, Keji He, and Liang Wang. 2024. Etpnav: Evolving topological planning for vision-language navigation in continuous environments.IEEE Transactions on Pattern Analysis and Machine Intelligence(2024)

    2024

  3. [3]

    Peter Anderson, Qi Wu, Damien Teney, Jake Bruce, Mark Johnson, Niko Sünder- hauf, Ian Reid, Stephen Gould, and Anton Van Den Hengel. 2018. Vision-and- language navigation: Interpreting visually-grounded navigation instructions in real environments. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 3674–3683

    work page 2018

  4. [4]

    Howard Chen, Alane Suhr, Dipendra Misra, Noah Snavely, and Yoav Artzi. 2019. Touchdown: Natural language navigation and spatial reasoning in visual street en- vironments. InProceedings of the IEEE/CVF International Conference on Computer Vision. 12538–12547

    work page 2019

  5. [5]

    Jinyu Chen, Chen Gao, Erli Meng, Qiong Zhang, and Si Liu. 2022. Reinforced structured state-evolution for vision-language navigation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 15450–15459

    2022

  6. [6]

    Runnan Chen, Youquan Liu, Lingdong Kong, Xinge Zhu, Yuexin Ma, Yikang Li, Yuenan Hou, Yu Qiao, and Wenping Wang. 2023. Clip2scene: Towards label- efficient 3d scene understanding by clip. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 7020–7030

    2023

  7. [7]

    Shizhe Chen, Pierre-Louis Guhur, Cordelia Schmid, and Ivan Laptev. 2021. History aware multimodal transformer for vision-and-language navigation.Advances in Neural Information Processing Systems34 (2021), 5834–5847

    work page 2021

  8. [8]

    Wei-Lin Chiang, Zhuohan Li, Ziqing Lin, Ying Sheng, Zhanghao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yonghao Zhuang, Joseph E Gonzalez, et al. 2023. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality. https://vicuna.lmsys.org. 2, 3 (2023), 6

    work page 2023

Show all 57 references
  1. [9]

    Hongyu Ding, Xinyue Liang, Yudong Fang, You Wu, Jieqi Shi, Jing Huo, Wenbin Li, Jing Wu, Yu-Kun Lai, and Yang Gao. 2025. SEA: Semantic Map Prediction for Active Exploration of Uncertain Areas.arXiv preprint arXiv:2510.19766(2025)

    2025

  2. [10]

    Daniel Fried, Ronghang Hu, Volkan Cirik, Anna Rohrbach, Jacob Andreas, Louis- Philippe Morency, Taylor Berg-Kirkpatrick, Kate Saenko, Dan Klein, and Trevor Darrell. 2018. Speaker-follower models for vision-and-language navigation. Advances in Neural Information Processing Syst...

    2018

  3. [11]

    Chen Gao, Jinyu Chen, Si Liu, Luting Wang, Qiong Zhang, and Qi Wu. 2021. Room-and-object aware knowledge reasoning for remote embodied referring expression. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 3064–3073

    2021

  4. [12]

    Chen Gao, Si Liu, Jinyu Chen, Luting Wang, Qi Wu, Bo Li, and Qi Tian. 2023. Room-object entity prompting and reasoning for embodied referring expression. IEEE Transactions on Pattern Analysis and Machine Intelligence46, 2 (2023), 994– 1010

    2023

  5. [13]

    Yunpeng Gao, Chenhui Li, Zhongrui You, Junli Liu, Zhen Li, Pengan Chen, Qizhi Chen, Zhonghan Tang, Liansheng Wang, Penghui Yang, et al. 2025. OpenFly: A comprehensive platform for aerial vision-language navigation.arXiv preprint arXiv:2502.18041(2025)

    2025

  6. [14]

    Yunpeng Gao, Zhigang Wang, Linglin Jing, Dong Wang, Xuelong Li, and Bin Zhao. 2024. Aerial vision-and-language navigation via semantic-topo-metric representation guided llm reasoning.arXiv preprint arXiv:2410.08500(2024)

    2024 arXiv

  7. [15]

    Georgios Georgakis, Karl Schmeckpeper, Karan Wanchoo, Soham Dan, Eleni Miltsakaki, Dan Roth, and Kostas Daniilidis. 2022. Cross-modal map learning for vision and language navigation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 15460–15470

    2022

  8. [16]

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick

  9. [17]

    Chenguang Huang, Oier Mees, Andy Zeng, and Wolfram Burgard. 2023. Visual language maps for robot navigation. In2023 IEEE International Conference on Robotics and Automation. IEEE, 10608–10615

    2023

  10. [18]

    Yanjia Huang, Mingyang Wu, Renjie Li, and Zhengzhong Tu. 2025. Vista: Gen- erative visual imagination for vision-and-language navigation.arXiv preprint arXiv:2505.07868(2025)

    2025

  11. [19]

    Vihan Jain, Gabriel Magalhaes, Alexander Ku, Ashish Vaswani, Eugene Ie, and Ja- son Baldridge. 2019. Stay on the path: Instruction fidelity in vision-and-language navigation. InProceedings of the 57th Annual Meeting of the Association for Com- putational Linguistics. 1862–1872

    2019

  12. [20]

    Jing Yu Koh, Honglak Lee, Yinfei Yang, Jason Baldridge, and Peter Anderson

  13. [21]

    Jacob Krantz, Erik Wijmans, Arjun Majumdar, Dhruv Batra, and Stefan Lee

  14. [22]

    Alexander Ku, Peter Anderson, Roma Patel, Eugene Ie, and Jason Baldridge. 2020. Room-across-room: Multilingual vision-and-language navigation with dense spatiotemporal grounding. InProceedings of the 2020 Conference on Empirical Methods in Natural Language Processing. 4392–4412

    2020

  15. [23]

    Jungdae Lee, Taiki Miyanishi, Shuhei Kurita, Koya Sakamoto, Daichi Azuma, Yutaka Matsuo, and Nakamasa Inoue. 2025. Citynav: A large-scale dataset for real- world aerial navigation. InProceedings of the IEEE/CVF International Conference on Computer Vision. 5912–5922

    2025

  16. [24]

    Jialu Li and Mohit Bansal. 2023. Improving vision-and-language navigation by generating future-view image semantics. InProceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition. 10803–10812

    2023

  17. [25]

    Yanwei Li, Chengyao Wang, and Jiaya Jia. 2024. Llama-vid: An image is worth 2 tokens in large language models. InProceedings of the European Conference on Computer Vision. 323–340

    2024

  18. [26]

    Guoxin Lian, Shuo Wang, Yucheng Wang, Yongcai Wang, Maiyue Chen, Kaihui Wang, Bo Zhang, Zhizhong Su, Deying Li, and Zhaoxin Fan. 2026. MapDream: Task-Driven Map Learning for Vision-Language Navigation.arXiv preprint arXiv:2602.00222(2026)

    2026

  19. [27]

    Bingqian Lin, Yunshuang Nie, Ziming Wei, Jiaqi Chen, Shikui Ma, Jianhua Han, Hang Xu, Xiaojun Chang, and Xiaodan Liang. 2025. Navcot: Boosting llm- based vision-and-language navigation via learning disentangled reasoning.IEEE Transactions on Pattern Analysis and Machine Intell...

    2025

  20. [28]

    Fei Liu, Shichao Xie, Minghua Luo, Zedong Chu, Junjun Hu, Xiaolong Wu, and Mu Xu. 2025. NavForesee: A Unified Vision-Language World Model for Hi- erarchical Planning and Dual-Horizon Navigation Prediction.arXiv preprint arXiv:2512.01550(2025)

    2025

  21. [29]

    Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Qing Jiang, Chunyuan Li, Jianwei Yang, Hang Su, et al. 2024. Grounding dino: Marrying dino with grounded pre-training for open-set object detection. InProceedings of the European Conference on Computer Visio...

    2024

  22. [30]

    Shubo Liu, Hongsheng Zhang, Yuankai Qi, Peng Wang, Yanning Zhang, and Qi Wu. 2023. Aerialvln: Vision-and-language navigation for uavs. InProceedings of the IEEE/CVF International Conference on Computer Vision. 15384–15394

    2023

  23. [31]

    Ilya Loshchilov and Frank Hutter. 2017. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101(2017)

    2017 arXiv

  24. [32]

    Laksh Nanwani, Kumaraditya Gupta, Aditya Mathur, Swayam Agrawal, AH Ab- dul Hafez, and K Madhava Krishna. 2024. Open-set 3D semantic instance maps for vision language navigation–O3D-SIM.Advanced Robotics38, 19-20 (2024), 1378–1391

    2024

  25. [33]

    Songyou Peng, Kyle Genova, Chiyu Jiang, Andrea Tagliasacchi, Marc Pollefeys, Thomas Funkhouser, et al. 2023. Openscene: 3d scene understanding with open vocabularies. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 815–824

    2023

  26. [34]

    Yuankai Qi, Qi Wu, Peter Anderson, Xin Wang, William Yang Wang, Chunhua Shen, and Anton van den Hengel. 2020. Reverie: Remote embodied visual referring expression in real indoor environments. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. ...

    2020

  27. [35]

    Hengcan Shi, Wen Liu, Zheng Li, Xinpu Fang, Xiangfeng Meng, Weixing Peng, Hang Zhong, Min Liu, and Yaonan Wang. 2026. Intelligent robot systems: a survey from the perspective of visual intelligence.Visual Intelligence4, 1 (2026), 14

    2026

  28. [36]

    Quan Sun, Yuxin Fang, Ledell Wu, Xinlong Wang, and Yue Cao. 2023. Eva-clip: Improved training techniques for clip at scale.arXiv preprint arXiv:2303.15389 (2023)

    2023 arXiv

  29. [37]

    Hanqing Wang, Wenguan Wang, Wei Liang, Caiming Xiong, and Jianbing Shen

  30. [38]

    Su Wang, Ceslee Montgomery, Jordi Orbay, Vighnesh Birodkar, Aleksandra Faust, Izzeddin Gur, Natasha Jaques, Austin Waters, Jason Baldridge, and Peter Anderson. 2022. Less is more: Generating grounded navigation instructions from landmarks. InProceedings of the IEEE/CVF Confere...

    2022

  31. [39]

    Xin Wang, Qiuyuan Huang, Asli Celikyilmaz, Jianfeng Gao, Dinghan Shen, Yuan- Fang Wang, William Yang Wang, and Lei Zhang. 2019. Reinforced cross-modal matching and self-supervised imitation learning for vision-language naviga- tion. InProceedings of the IEEE/CVF Conference on ...

    2019

  32. [40]

    Xiangyu Wang, Donglin Yang, Ziqin Wang, Hohin Kwan, Jinyu Chen, Wenjun Wu, Hongsheng Li, Yue Liao, and Si Liu. 2024. Towards realistic uav vision- language navigation: Platform, benchmark, and methodology.arXiv preprint arXiv:2410.07087(2024)

    2024 arXiv

  33. [41]

    InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Structured scene memory for vision-language navigation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 8455–8464

  34. [42]

    Xuan Yao, Junyu Gao, and Changsheng Xu. 2025. Navmorph: A self-evolving world model for vision-and-language navigation in continuous environments. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 5536– 5546. MM ’26, November 10–14, 2026, Rio de Janei...

    2025

  35. [43]

    Naoki Yokoyama, Sehoon Ha, Dhruv Batra, Jiuguang Wang, and Bernadette Bucher. 2024. Vlfm: Vision-language frontier maps for zero-shot semantic navi- gation. In2024 IEEE International Conference on Robotics and Automation. IEEE, 42–48

    2024

  36. [44]

    Chaoning Zhang, Dongshen Han, Yu Qiao, Jung Uk Kim, Sung-Ho Bae, Seungkyu Lee, and Choong Seon Hong. 2023. Faster segment anything: Towards lightweight sam for mobile applications.arXiv preprint arXiv:2306.14289(2023)

    2023 arXiv

  37. [45]

    Peiran Xu, Xicheng Gong, and Yadong Mu. 2025. NavQ: Learning a Q-Model for Foresighted Vision-and-Language Navigation. InProceedings of the IEEE/CVF International Conference on Computer Vision. 6327–6341

    2025

  38. [46]

    Lingfeng Zhang, Xiaoshuai Hao, Qinwen Xu, Qiang Zhang, Xinyao Zhang, Peng- wei Wang, Jing Zhang, Zhongyuan Wang, Shanghang Zhang, and Renjing Xu

  39. [47]

    Weichen Zhang, Chen Gao, Shiquan Yu, Ruiying Peng, Baining Zhao, Qian Zhang, Jinqiang Cui, Xinlei Chen, and Yong Li. 2025. Citynavagent: Aerial vision-and- language navigation with hierarchical semantic planning and global memory. InProceedings of the 63rd Annual Meeting of th...

    2025

  40. [48]

    Wenyao Zhang, Hongsi Liu, Zekun Qi, Yunnan Wang, Xinqiang Yu, Jiazhao Zhang, Runpei Dong, Jiawei He, Fan Lu, He Wang, et al . 2025. Dreamvla: a vision-language-action model dreamed with comprehensive world knowledge. arXiv preprint arXiv:2507.04447(2025)

    2025 arXiv

  41. [49]

    Jiazhao Zhang, Anqi Li, Yunpeng Qi, Minghan Li, Jiahang Liu, Shaoan Wang, Haoran Liu, Gengze Zhou, Yuze Wu, Xingxing Li, et al. 2025. Embodied navigation foundation model.arXiv preprint arXiv:2509.12129(2025)

    2025

  42. [50]

    Ganlong Zhao, Guanbin Li, Jia Pan, and Yizhou Yu. 2025. Aerial vision-and- language navigation with grid-based view selection and map construction.arXiv preprint arXiv:2503.11091(2025)

    2025 arXiv

  43. [51]

    Qingqing Zhao, Yao Lu, Moo Jin Kim, Zipeng Fu, Zhuoyang Zhang, Yecheng Wu, Zhaoshuo Li, Qianli Ma, Song Han, Chelsea Finn, et al . 2025. Cot-vla: Visual chain-of-thought reasoning for vision-language-action models. InProceedings of the IEEE/CVF International Conference on Comp...

    2025

  44. [52]

    Step 1 Step 2 Step 5 Step 9 Step 17 Step 21 Step 24 Step 2 Step 3 Step 4 Step 6 Step 9 Step 10 AirForesight Step 1 TravelUAV Object Description:

    Yusheng Zhao, Jinyu Chen, Chen Gao, Wenguan Wang, Lirong Yang, Haibing Ren, Huaxia Xia, and Si Liu. 2022. Target-driven structured transformer planner for vision-language navigation. InProceedings of the 30th ACM International Conference on Multimedia. 4194–4203. A Implementat...

    2022

  45. [54]

    Yue Zhang and Parisa Kordjamshidi. 2023. Vln-trans: Translator for the vision and language navigation agent. InProceedings of the 61st Annual Meeting of the Association for Computational Linguistics. 13219–13233

    2023

  46. [2020]

    InProceedings of the European Conference on Computer Vision

    Beyond the nav-graph: Vision-and-language navigation in continuous environments. InProceedings of the European Conference on Computer Vision. Springer, 104–120

  47. [2021]

    InProceedings of the IEEE/CVF International Conference on Computer Vision

    Pathdreamer: A world model for indoor navigation. InProceedings of the IEEE/CVF International Conference on Computer Vision. 14738–14748

  48. [2022]

    InProceedings of the IEEE/CVF International Conference on Computer Vision

    Masked autoencoders are scalable vision learners. InProceedings of the IEEE/CVF International Conference on Computer Vision. 16000–16009

  49. [2025]

    InProceedings of the 63rd Annual Meeting of the Association for Computational Linguistics

    Mapnav: A novel memory representation via annotated semantic maps for vlm-based vision-and-language navigation. InProceedings of the 63rd Annual Meeting of the Association for Computational Linguistics. 13032–13056

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.