Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

GC-VLN: Instruction as Graph Constraints for Training-free Vision-and-Language Navigation

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

Pith's one-line read A robot can follow natural-language directions with zero training by turning them into spatial constraints and solving them.

desk verdict A genuinely new graph-constraint mechanism for training-free VLN-CE with strong RxR results, but the efficiency claim is contradicted by SPL and the constraint-library completeness is asserted rather than demonstrated. read the letter →

arxiv 2509.10454 v1 pith:PAZ6EZFX submitted 2025-09-12 cs.RO cs.CV

classification cs.ROcs.CV
keywords vision-and-languagenavigationzero-shottraining-freegraphconstraintsconstraintoptimizationcontinuousenvironmentstree
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

GC-VLN claims that a robot can follow natural-language navigation instructions in unseen continuous environments with no training at all. The method reduces each instruction to a directed acyclic graph of waypoint and object nodes, matches every spatial relation against a six-template constraint library, and solves the resulting constraint optimization to locate the waypoints one stage at a time. A navigation tree stores alternate solutions, so when one interpretation of an instruction leads to a dead end, the robot backtracks to a different solution rather than stopping. On the R2R-CE and RxR-CE benchmarks it reports higher success rate and navigation efficiency than previous zero-shot VLN-CE methods, and the same pipeline runs on a real robot. If correct, this turns instruction following from a data-hungry learning problem into a perception-plus-constraint-solving problem.

What carries the argument

$K=(V,C)$, the graph constraint formed from the instruction's directed acyclic graph and the six-type spatial constraint library, is the mechanism that carries the argument. The library's templates cover front/back relations and relations like 'through a door' or 'between two objects', each contributing an angle subconstraint and/or a distance subconstraint with defaults $\Delta\phi=45^\circ$, $d=1.5\mathrm{m}$, $\Delta d=1.5\mathrm{m}$. A topological sort fixes the order in which node coordinates are solved; a constrained optimizer places each waypoint by maximizing the sum of subconstraints subject to every subconstraint being nonnegative; and the navigation tree stores all alternative waypoint solutions so that a failed branch triggers backtracking instead of ending the episode.

What would settle it

Construct a set of R2R-CE or real-world instructions built around spatial relations that are absent from the six templates, such as 'follow the curved corridor around to the right' or 'stay on the left side of the room as you cross it', and run the full system on them. If the LLM cannot decompose them into the allowed object-position categories, or the success rate falls far below the reported 33.6% SR, then the library does not cover all VLN spatial relationships as claimed. A cheaper quantitative check is to sweep $\Delta\phi$ and $d$ over a wide range and see whether the defaults of 45 degrees and 1.5 meters are genuinely robust or merely tuned to the benchmark.

Watch

Extended reading notes

Core claim

The central claim is that every VLN instruction can be encoded as a graph constraint problem whose solution is the path, with no learned navigation policy. The instruction is parsed by an LLM into stages, each containing exactly one displacement and the objects encountered; those objects and waypoints form nodes, and the spatial relations between them form edges typed by the constraint library. Waypoint coordinates are then solved sequentially in topological order by a nonlinear constrained optimizer, while object coordinates come from projecting pretrained-vision detections onto a bird's-eye-view map. When a node has multiple feasible coordinates, the solver branches; when no coordinate satisfies the constraints, the robot backtracks to an unexplored branch. The paper presents this constraint-solving pipeline as the reason it reaches state-of-the-art zero-shot success on both benchmarks and generalizes to a physical robot.

Load-bearing premise

The framework stands on the claim that every spatial relationship in VLN instructions falls into one of six templates with default tolerances of 45 degrees and 1.5 meters; if some instruction's relation does not fit, the resulting graph misrepresents the path and navigation fails.

Editorial extensions

If this is right

  • If the claim is right, zero-shot VLN-CE no longer needs simulator self-supervised training, so deploying to a new building only requires perception and an LLM.
  • The same constraint graph can be reused with a different local planner or a different object detector, since the navigation decision is made by the solver, not by a learned policy.
  • The navigation tree gives the planner a built-in recovery mechanism: misperceived objects or wrong turns are handled by exhausting alternative waypoint solutions before declaring failure.
  • Because the pipeline is training-free, adding a new language to the benchmark only requires the LLM to parse it, not new navigation data.

Reading between the lines

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

  • The six-template library can be read as a claim about the grammar of navigation instructions; a natural test is to count how often each preposition in VLN corpora maps to a template and where the residue lies.
  • The fixed angle and distance tolerances suggest the approach is best suited to indoor spaces; outdoors or in large open environments, the default $d=1.5\mathrm{m}$ would likely need to scale with environment size, a modification the paper does not explore.
  • The same graph-constraint formulation could be applied to instruction-following tasks outside navigation, such as mobile manipulation, by treating object affordances as additional constraint templates.
  • The paper reports zero-shot results but depends on pretrained components such as the LLM and vision model; separating how much of the gain comes from the constraint solver itself versus the choice of those components would sharpen the attribution of the improvement.
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. GC-VLN proposes a training-free framework for vision-and-language navigation in continuous environments (VLN-CE). The method decomposes a natural-language instruction into a directed acyclic graph of waypoint and object nodes, queries a hand-built library of six spatial constraint types to convert the graph into a set of geometric constraints, and solves a constrained optimization problem to determine waypoint coordinates. A navigation tree with backtracking handles cases where the constraint solver yields no or multiple solutions. The paper reports experiments on the R2R-CE and RxR-CE benchmarks, claiming state-of-the-art zero-shot success rates, and presents qualitative real-world deployments. The central claim is that the constraint library covers all spatial relationships found in VLN instructions, making the approach general and training-free.

Significance. If the claims hold, GC-VLN is a conceptually interesting approach that replaces trained navigation policies with explicit spatial constraint solving, potentially enabling deployment in unseen environments without collecting task-specific training data. The paper includes extensive simulator experiments, ablations of the pipeline components, and real-world robot demonstrations, which are valuable. The method achieves competitive SR on R2R-CE and strong results on RxR-CE relative to reported zero-shot baselines. However, the stated navigation-efficiency improvement is contradicted by the paper's own SPL numbers, and the completeness assertion for the constraint library is not substantiated. These issues temper the significance unless they are addressed.

major comments (4)
  1. [§1, §4.2, Table 1] The abstract and Section 4.2 claim 'significant improvements in success rate and navigation efficiency' compared to state-of-the-art zero-shot VLN methods, but Table 1 shows that on R2R-CE the proposed method's SPL (16.3) is substantially lower than the training-free baseline InstructNav (24.0), despite a higher SR (33.6 vs. 31.0). Since SPL is the standard metric for navigation efficiency, the paper's efficiency claim is not supported; the narrative selectively emphasizes SR. Please either revise the claim, report SPL alongside SR in all comparisons, or justify an alternative efficiency metric.
  2. [§3.2, Supplementary D.1] The paper asserts that the six-type constraint library 'covers all types of spatial relationship topologies involved in VLN instructions,' but this is not demonstrated. Supplementary Section D.1 lists seven allowed 'object position' labels ('right', 'left', 'through', 'weave', 'pass', 'near', 'back') without an explicit mapping to the six constraint types, and no coverage analysis is provided for the actual instructions in R2R-CE or RxR-CE. Common relational phrases such as 'behind the table' or 'past the staircase' are not shown to be representable. Because the entire pipeline depends on this coverage, please provide a mapping from instruction phrases to constraint types and an audit of how the constraint library covers the relations appearing in the benchmark datasets.
  3. [§3.2, Supplementary D.2, Table 3] The default tolerances Δφ=45°, d=1.5 m, and Δd=1.5 m are not derived from the instruction or the environment but are selected through validation-set ablations reported in Table 3 of the supplementary material. This weakens the 'training-free' characterization and raises a generalization concern: Table 3 shows SR varying from 28.5 to 33.6 as these hyperparameters change, so the defaults are not robust across plausible settings. Please justify the defaults by analyzing their effect across environments or by deriving them from dataset statistics, and discuss how they would transfer to new scenes with different scales.
  4. [§3.3, Eqs. (1)-(3)] The constraint formulations do not appear to implement exactly the stated angle and distance tolerances. For constraint type 2, the condition c_a ≥ 0 with equation (2) is equivalent to cos(δ) ≥ 1 - cos(Δφ), where δ is the angle deviation, rather than simply δ ≤ Δφ; for Δφ=45° this yields an effective angular bound of roughly 73°, not 45°. Since the graph constraints are the core of the method, please clarify the intended semantics of Δφ and either correct the sub-constraint definitions so that the tolerance is enforced exactly or explicitly state that the constraints are soft approximations.
minor comments (6)
  1. [Table 1] Some entries are marked with '-' without explaining whether the method was not evaluated or the metric was not reported; please add a note clarifying the meaning of dashes.
  2. [§4.2] The comparison with InstructNav is incomplete: the paper reports only the SR gap and omits the SPL, which is the metric that would support or refute the navigation-efficiency claim; please present a balanced comparison including SPL and NE.
  3. [Figures 2 and 4] The notation such as 't.i' in the graph and navigation-tree figures is not defined in the main text; please define it in the caption or in Section 3.3.
  4. [§4.4] The real-world experiments are described qualitatively; please include quantitative success metrics (e.g., number of trials and success rate) to substantiate the claimed generalization ability.
  5. [Throughout] There are several typos, e.g., 'Consrtraint' in the captions of Tables 2 and 3, 'objest' in Section 2.2, and 'ojbect-goal navigation' in Section 4.1; please proofread the manuscript.
  6. [References] The 'Project Page' link in the abstract is not included in the text or references; please provide a URL or remove the reference.

Circularity Check

0 steps flagged · score 0.0 of 10

No load-bearing circularity: the constraint-solving pipeline is evaluated against external benchmarks and its equations are not restatements of the success metric.

full rationale

GC-VLN's claimed derivation—decompose the instruction into a DAG, retrieve constraints from the six-type library, solve for waypoint coordinates, and navigate—does not reduce to its inputs at the equation level. The sub-constraints c_a and c_d in Eqs. (2)-(3) and the supplement's Eqs. (8)-(16) are functions of instruction-specified angle and distance baselines together with object positions perceived from RGB-D observations; the optimization objective sum(c)/min(c) is not a restatement of SR/SPL, and neither the ground-truth endpoint nor the success radius r=3 m enters the constraint construction. The library-completeness claim ('covers all types of spatial relationship topologies involved in VLN instructions', Sec. 3.2) is an unproven coverage assumption rather than a definitional identity: the prompt's allowed output labels do not make natural-language coverage true by construction, and any mismatch is a correctness risk, not a circular derivation. The authors' self-citations (SG-Nav, UniGoal, IGL-Nav) appear in related work only and are not load-bearing. The main caveat is that the default tolerances (Δφ=45°, d=1.5 m, Δd=1.5 m) are selected via validation-set ablations (Supp. Table 3) on the same R2R-CE validation-unseen split that is later reported, which weakens the 'training-free' framing; however, this is benchmark hyperparameter tuning, not a case where the reported performance is statistically forced by the fit or where the derivation is equivalent to its inputs. No circular step meets the evidence bar.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The framework's load-bearing components are hand-constructed: a six-template constraint library, fixed default angle/distance tolerances, and a hand-designed optimization objective. These are not derived from data or first principles, and the supplied ablations only sample a few values. No new physical entities are introduced.

free parameters (3)
  • Angle tolerance Δφ = 45°
    Default for constraint types 1, 2, 3, and 6; selected by validation ablation in Supplementary Table 3, where 30°, 45°, and 75° are compared and 45° gives the best SR.
  • Distance baseline d = 1.5 m
    Default distance baseline for all distance constraints; ablations in Supplementary Table 3 compare 0.8 m, 1.5 m, 2.5 m and 1.5 m is chosen.
  • Distance tolerance Δd = 1.5 m
    Default tolerance for distance constraints; no ablation is reported, the value is hand-set.
assumptions (5)
  • domain assumption Every navigation instruction can be decomposed into stages, each containing exactly one displacement, with waypoint and object nodes forming a DAG.
    Section 3.2 'Instruction Decomposition' states this as a rule for the LLM; if an instruction involves multiple moves or objects that span stages ambiguously, the graph becomes incorrect.
  • ad hoc to paper The six constraint types in the library cover all spatial relationships in VLN instructions.
    Section 3.2 and Supplementary D.2 assert completeness without proof or a formal taxonomy; the seven object-position labels are enumerated by hand.
  • domain assumption Pre-trained models (Grounding DINO/Grounded-SAM-2) can reliably detect all objects mentioned in the instruction and project them to the BEV map.
    Section 4.1 lists Grounded-SAM-2 as the perception model; failure cases in Figure 7 include 'fails to locate the globe', showing this assumption is not always met.
  • ad hoc to paper The nonlinear constrained optimization (P1) with the sum-of-constraints objective yields a feasible and navigationally correct waypoint for any satisfiable constraint set.
    Section 3.3 defines (P1) as maximization of sum(c) subject to min(c) ≥ 0 and separation; there is no guarantee this objective matches human navigation semantics.
  • standard math The local Fast Marching Method policy can execute the waypoint-to-waypoint motion in the environment.
    Section 4.1 states the local policy is FMM; this is a standard path-planning method and is a reasonable but unverified assumption in cluttered real scenes.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GC-VLN: Instruction as Graph Constraints for Training-free Vision-and-Language Navigation." pith.science (2026). https://pith.science/paper/PAZ6EZFX

@misc{pith2026250910454,
  author       = {Pith},
  title        = {Pith review of: GC-VLN: Instruction as Graph Constraints for Training-free Vision-and-Language Navigation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PAZ6EZFX}},
  note         = {Machine review of arXiv:2509.10454}
}
read the original abstract

In this paper, we propose a training-free framework for vision-and-language navigation (VLN). Existing zero-shot VLN methods are mainly designed for discrete environments or involve unsupervised training in continuous simulator environments, which makes it challenging to generalize and deploy them in real-world scenarios. To achieve a training-free framework in continuous environments, our framework formulates navigation guidance as graph constraint optimization by decomposing instructions into explicit spatial constraints. The constraint-driven paradigm decodes spatial semantics through constraint solving, enabling zero-shot adaptation to unseen environments. Specifically, we construct a spatial constraint library covering all types of spatial relationship mentioned in VLN instructions. The human instruction is decomposed into a directed acyclic graph, with waypoint nodes, object nodes and edges, which are used as queries to retrieve the library to build the graph constraints. The graph constraint optimization is solved by the constraint solver to determine the positions of waypoints, obtaining the robot's navigation path and final goal. To handle cases of no solution or multiple solutions, we construct a navigation tree and the backtracking mechanism. Extensive experiments on standard benchmarks demonstrate significant improvements in success rate and navigation efficiency compared to state-of-the-art zero-shot VLN methods. We further conduct real-world experiments to show that our framework can effectively generalize to new environments and instruction sets, paving the way for a more robust and autonomous navigation framework.

Figures

Figures reproduced from arXiv: 2509.10454 by the authors.

Figure 1
Figure 1. GC-VLN models the instructions as a graph constraint optimization problem and solve [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Framework of GC-VLN. We construct a constraint library, containing all the spatial rela [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Diagram of the Constraint Library con￾taining six types of constraint. For constraint c (v | u), u and v are colored blue and green, re￾spectively. The green region is the possible region for v. To handle the long sequence characteristics and complex spatial relationships in the instruction I, we convert I into a structured representation, namely the graph constraint, which is required to meet three criteria: 1. It … view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Demonstration of the graph constraints solving of GC-VLN. Here [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Demonstration of deployment in real-world environment. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: The robot employed for conducting the real-world experimental deployments. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Visualization of three failure cases. In the first case, robot fails to locate the globe. In [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Rule-VLN: Bridging Perception and Compliance via Semantic Reasoning and Geometric Rectification

    cs.AI 2026-04 unverdicted novelty 7.0 of 10

    Rule-VLN is the first large-scale benchmark injecting 177 regulatory categories into an urban environment, and the proposed SNRM module equips pre-trained VLN agents with zero-shot semantic reasoning and detour planni...

Reference graph

Works this paper leans on

50 extracted references · 25 canonical work pages · cited by 1 Pith paper

  1. [1]

    Anderson, Q

    P. Anderson, Q. Wu, D. Teney, J. Bruce, M. Johnson, N. S ¨underhauf, I. Reid, S. Gould, and A. van den Hengel. Vision-and-language navigation: Interpreting visually-grounded naviga- tion instructions in real environments. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018

  2. [2]

    D. An, H. Wang, W. Wang, Z. Wang, Y . Huang, K. He, and L. Wang. Etpnav: Evolving topological planning for vision-language navigation in continuous environments.IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 2024

  3. [3]

    P. Chen, D. Ji, K. Lin, R. Zeng, T. H. Li, M. Tan, and C. Gan. Weakly- supervised multi-granularity map learning for vision-and-language navigation.arXiv preprint arXiv:2210.07506, 2022

  4. [4]

    Zhang, K

    J. Zhang, K. Wang, R. Xu, G. Zhou, Y . Hong, X. Fang, Q. Wu, Z. Zhang, and H. Wang. Navid: Video-based vlm plans the next step for vision-and-language navigation.Robotics: Science and Systems, 2024

  5. [5]

    G. Zhou, Y . Hong, and Q. Wu. Navgpt: Explicit reasoning in vision-and-language navigation with large language models.arXiv preprint arXiv:2305.16986, 2023

  6. [6]

    J. Chen, B. Lin, R. Xu, Z. Chai, X. Liang, and K.-Y . K. Wong. Mapgpt: Map-guided prompting with adaptive path planning for vision-and-language navigation. InProceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, 2024

  7. [7]

    Y . Long, X. Li, W. Cai, and H. Dong. Discuss before moving: Visual language navigation via multi-expert discussions. In2024 IEEE International Conference on Robotics and Automation (ICRA), pages 17380–17387. IEEE, 2024

  8. [8]

    Krantz, E

    J. Krantz, E. Wijmans, A. Majundar, D. Batra, and S. Lee. Beyond the nav-graph: Vision and language navigation in continuous environments. InEuropean Conference on Computer Vision (ECCV), 2020

Show all 50 references
  1. [9]

    P. Chen, X. Sun, H. Zhi, R. Zeng, T. H. Li, G. Liu, M. Tan, and C. Gan.a 2nav: Action-aware zero-shot robot navigation by exploiting vision-and-language ability of foundation models,

  2. [10]

    W. Hao, C. Li, X. Li, L. Carin, and J. Gao. Towards learning a generic agent for vision-and- language navigation via pre-training.Conference on Computer Vision and Pattern Recognition (CVPR), 2020

  3. [11]

    Anderson, A

    P. Anderson, A. Shrivastava, J. Truong, A. Majumdar, D. Parikh, D. Batra, and S. Lee. Sim- to-real transfer for vision-and-language navigation. InConference on Robot Learning, pages 671–681. PMLR, 2021

  4. [12]

    K. Chen, J. K. Chen, J. Chuang, M. V ´azquez, and S. Savarese. Topological planning with transformers for vision-and-language navigation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11276–11286, 2021

  5. [13]

    R. Liu, X. Wang, W. Wang, and Y . Yang. Bird’s-eye-view scene graph for vision-language navigation. InICCV, pages 10968–10980, 2023. 9

  6. [14]

    Krantz and S

    J. Krantz and S. Lee. Sim-2-sim transfer for vision-and-language navigation in continuous environments. InEuropean conference on computer vision, pages 588–603. Springer, 2022

  7. [15]

    H. Wang, W. Liang, L. Van Gool, and W. Wang. Dreamwalker: Mental planning for continu- ous vision-language navigation. InProceedings of the IEEE/CVF international conference on computer vision, pages 10873–10883, 2023

  8. [16]

    Z. Wang, X. Li, J. Yang, Y . Liu, and S. Jiang. Gridmm: Grid memory map for vision-and- language navigation. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 15625–15636, 2023

  9. [17]

    Z. Wang, X. Li, J. Yang, Y . Liu, J. Hu, M. Jiang, and S. Jiang. Lookahead exploration with neu- ral radiance representation for continuous vision-language navigation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 13753– 1376...

  10. [18]

    Y . Hong, Z. Wang, Q. Wu, and S. Gould. Bridging the gap between learning in discrete and continuous environments for vision-and-language navigation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2022

  11. [19]

    Hochreiter and J

    S. Hochreiter and J. Schmidhuber. Long short-term memory.Neural computation, 9(8):1735– 1780, 1997

  12. [20]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polo- sukhin. Attention is all you need.Advances in neural information processing systems, 30, 2017

  13. [21]

    B. Yu, H. Kasaei, and M. Cao. L3mvn: Leveraging large language models for visual target navigation. In2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 3554–3560. IEEE, 2023

  14. [22]

    H. Yin, X. Xu, Z. Wu, J. Zhou, and J. Lu. Sg-nav: Online 3d scene graph prompting for llm-based zero-shot object navigation.arXiv preprint arXiv:2410.08189, 2024

  15. [23]

    K. Zhou, K. Zheng, C. Pryor, Y . Shen, H. Jin, L. Getoor, and X. E. Wang. Esc: Exploration with soft commonsense constraints for zero-shot object navigation. InICML, pages 42829–42842. PMLR, 2023

  16. [24]

    W. Cai, S. Huang, G. Cheng, Y . Long, P. Gao, C. Sun, and H. Dong. Bridging zero-shot object navigation and foundation models through pixel-guided navigation skill, 2023. URL https://arxiv.org/abs/2309.10309

  17. [25]

    P. Wu, Y . Mu, B. Wu, Y . Hou, J. Ma, S. Zhang, and C. Liu. V oronav: V oronoi-based zero-shot object navigation with large language model.arXiv preprint arXiv:2401.02695, 2024

  18. [26]

    Kuang, H

    Y . Kuang, H. Lin, and M. Jiang. Openfmnav: Towards open-set zero-shot object navigation via vision-language foundation models.arXiv preprint arXiv:2402.10670, 2024

  19. [27]

    W. Guo, X. Xu, H. Yin, Z. Wang, J. Feng, J. Zhou, and J. Lu. Igl-nav: Incremental 3d gaussian localization for image-goal navigation.arXiv preprint arXiv:2508.00823, 2025

  20. [28]

    Krantz, T

    J. Krantz, T. Gervet, K. Yadav, A. Wang, C. Paxton, R. Mottaghi, D. Batra, J. Malik, S. Lee, and D. S. Chaplot. Navigating to objects specified by images.arXiv preprint arXiv:2304.01192, 2023

  21. [29]

    H. Yin, X. Xu, L. Zhao, Z. Wang, J. Zhou, and J. Lu. Unigoal: Towards universal zero-shot goal-oriented navigation.arXiv preprint arXiv:2503.10630, 2025

  22. [30]

    M. Wei, T. Wang, Y . Chen, H. Wang, J. Pang, and X. Liu. Ovexp: Open vocabulary exploration for object-oriented navigation.arXiv preprint arXiv:2407.09016, 2024. 10

  23. [31]

    X. Sun, L. Liu, H. Zhi, R. Qiu, and J. Liang. Prioritized semantic learning for zero-shot instance navigation, 2024. URLhttps://arxiv.org/abs/2403.11650

  24. [32]

    D. Li, W. Chen, and X. Lin. Tina: Think, interaction, and action framework for zero-shot vision language navigation, 2024. URLhttps://arxiv.org/abs/2403.08833

  25. [33]

    Zhang, K

    J. Zhang, K. Wang, S. Wang, M. Li, H. Liu, S. Wei, Z. Wang, Z. Zhang, and H. Wang. Uni- navid: A video-based vision-language-action model for unifying embodied navigation tasks, 2024

  26. [34]

    S. Y . Gadre, M. Wortsman, G. Ilharco, L. Schmidt, and S. Song. Cows on pasture: Baselines and benchmarks for language-driven zero-shot object navigation.CVPR, 2023

  27. [35]

    Majumdar, G

    A. Majumdar, G. Aggarwal, B. Devnani, J. Hoffman, and D. Batra. Zson: Zero-shot object- goal navigation using multimodal goal embeddings. InNeural Information Processing Systems (NeurIPS), 2022

  28. [36]

    K. Chen, D. An, Y . Huang, R. Xu, Y . Su, Y . Ling, I. Reid, and L. Wang. Constraint- aware zero-shot vision-language navigation in continuous environments.arXiv preprint arXiv:2412.10137, 2024

  29. [37]

    Y . Long, W. Cai, H. Wang, G. Zhan, and H. Dong. Instructnav: Zero-shot system for generic instruction navigation in unexplored environment, 2024

  30. [38]

    Anderson, Q

    P. Anderson, Q. Wu, D. Teney, J. Bruce, M. Johnson, N. S ¨underhauf, I. Reid, S. Gould, and A. Van Den Hengel. Vision-and-language navigation: Interpreting visually-grounded naviga- tion instructions in real environments. InProceedings of the IEEE conference on computer vision...

  31. [39]

    A. Ku, P. Anderson, R. Patel, E. Ie, and J. Baldridge. Room-across-room: Multilin- gual vision-and-language navigation with dense spatiotemporal grounding.arXiv preprint arXiv:2010.07954, 2020

  32. [40]

    Savva, A

    M. Savva, A. Kadian, O. Maksymets, Y . Zhao, E. Wijmans, B. Jain, J. Straub, J. Liu, V . Koltun, J. Malik, et al. Habitat: A platform for embodied ai research. InProceedings of the IEEE/CVF international conference on computer vision, pages 9339–9347, 2019

  33. [41]

    Chang, A

    A. Chang, A. Dai, T. Funkhouser, M. Halber, M. Niessner, M. Savva, S. Song, A. Zeng, and Y . Zhang. Matterport3d: Learning from rgb-d data in indoor environments.3DV, 2017

  34. [42]

    Anderson, A

    P. Anderson, A. Chang, D. S. Chaplot, A. Dosovitskiy, S. Gupta, V . Koltun, J. Kosecka, J. Ma- lik, R. Mottaghi, M. Savva, et al. On evaluation of embodied navigation agents.arXiv preprint arXiv:1807.06757, 2018

  35. [43]

    Ilharco, V

    G. Ilharco, V . Jain, A. Ku, E. Ie, and J. Baldridge. General evaluation for instruction condi- tioned navigation using dynamic time warping.arXiv preprint arXiv:1907.05446, 2019

  36. [44]

    G. Zhou, Y . Hong, Z. Wang, X. E. Wang, and Q. Wu. Navgpt-2: Unleashing navigational reasoning capability for large vision-language models.arXiv preprint arXiv:2407.12366, 2024

  37. [45]

    Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning, 2025

    DeepSeek-AI. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning, 2025. URLhttps://arxiv.org/abs/2501.12948

  38. [46]

    T. Ren, S. Liu, A. Zeng, J. Lin, K. Li, H. Cao, J. Chen, X. Huang, Y . Chen, F. Yan, Z. Zeng, H. Zhang, F. Li, J. Yang, H. Li, Q. Jiang, and L. Zhang. Grounded sam: Assembling open- world models for diverse visual tasks, 2024

  39. [47]

    N. Ravi, V . Gabeur, Y .-T. Hu, R. Hu, C. Ryali, T. Ma, H. Khedr, R. R ¨adle, C. Rolland, L. Gustafson, E. Mintun, J. Pan, K. V . Alwala, N. Carion, C.-Y . Wu, R. Girshick, P. Doll ´ar, and C. Feichtenhofer. Sam 2: Segment anything in images and videos, 2024. URLhttps: //arxiv...

  40. [48]

    T. Ren, Q. Jiang, S. Liu, Z. Zeng, W. Liu, H. Gao, H. Huang, Z. Ma, X. Jiang, Y . Chen, Y . Xiong, H. Zhang, F. Li, P. Tang, K. Yu, and L. Zhang. Grounding dino 1.5: Advance the ”edge” of open-set object detection, 2024

  41. [49]

    stage 1":{

    J. A. Sethian. A fast marching level set method for monotonically advancing fronts.proceed- ings of the National Academy of Sciences, 93(4):1591–1595, 1996. 12 A Overview This supplementary material is organized as follows: • Section B provides the algorithm for the overall pi...

  42. [2023]

    URLhttps://arxiv.org/abs/2308.07997

Pith tools

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