Pith. sign in

REVIEW 3 major objections 6 minor 40 references

HumanoidVLN: A Physics-Grounded Simulator and Benchmark for Vision-Language Navigation Across Diverse Humanoid Embodiments

T0 review · 3 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 HumanoidVLN replaces kinematic teleportation with full bipedal physics, creating a 933-episode vision-language navigation benchmark in which every scored trajectory is physically executable by the robot under test, and shows that model…

desk verdict A solid new humanoid VLN benchmark with real zero-shot evaluation and an honest pilot sim-to-real check; the main risk is that three of four embodiments are unvalidated, so the dramatic H1 fall rates may be simulation artifacts. read the letter →

arxiv 2608.12860 v1 pith:AAPPTBU2 submitted 2026-08-13 cs.RO

Quan-Dung Pham , Anh Dao , The-Anh Nguyen , Minh Nguyen-Dinh , Phuong Nam Dang , Tri Pham , Hung Tran , Bach Dao
show 3 more authors
Tuyen P. Le Truong Nguyen Quan Nguyen
This is my paper · ORCID
classification cs.RO
keywords vision-languagenavigationhumanoidrobotsphysicssimulationembodiedAIsim-to-realtransfer3DGaussianSplattingleglocomotionbenchmark
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 aims to prove that vision-language navigation (VLN) for humanoid robots — instructing a walking robot to follow natural-language directions — cannot be judged fairly on simulators that move the agent by teleporting it along a path, because a bipedal body changes which paths are passable, how much the camera shakes, and whether the robot falls. To test that claim, it builds a physics-based simulation and a 933-episode benchmark in which four humanoid robots, spanning heights of $1.17$–$1.80$ m and 10–12 lower-body degrees of freedom, execute every episode under a hierarchical controller that respects joint limits and gait stability. The paper reports that replacing kinematic stepping with real dynamics reshuffles the rankings of four VLN models across embodiments, with success rates from about 14% to 50% depending on the robot, and that a 20-episode sim-to-real pilot on one robot shows endpoint errors correlating at $r = 0.935$ with a mean absolute gap of $0.68$ m. If this holds, the benchmark offers a way to select both navigation models and robot morphologies before hardware deployment.

What carries the argument

The load-bearing object is the two-level control hierarchy that replaces teleportation: a per-embodiment reinforcement-learning locomotion policy, which commands joint torques under joint limits and center-of-mass constraints, driven by interchangeable high-level path trackers (proportional–derivative control for discrete-action models, model-predictive control for continuous-action models). This stack is what the paper invokes to guarantee that every evaluated trajectory is physically executable by the specific robot under test. Around it, two supporting mechanisms carry the benchmark: a scene pipeline that reconstructs real spaces with 3D Gaussian splatting, enforces depth–normal consistency and unbiased depth, fuses collision meshes, and keeps only scenes with at least $100\,\mathrm{m}^2$ of navigable floor; and a multi-agent instruction pipeline in which two generators independently construct route graphs from egocentric video, a reviewer verifies them against trajectory and scene-graph priors, a paraphraser produces three style variants, and humans correct the result.

What would settle it

Run the same evaluation episodes on the physical versions of the three robots that were never validated against reality; if their per-episode endpoint errors and fall events do not align with simulation as closely as the pilot's $r = 0.935$ and $0.68$ m gap, then the guarantee of physical executability does not transfer across embodiments.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that physical executability is not a detail of humanoid VLN but a first-order determinant of measured performance. By running four VLN models zero-shot on identical 933 episodes across four humanoid bodies, the paper finds that model and embodiment interact strongly: the model with explicit 3D spatial memory reaches the highest mean success rate ($43.55\%$) and path fidelity (normalized dynamic time warping, nDTW, of $48.38$), while the tallest, 10-DoF robot lowers average success rates to about $21\%$ and produces fall rates above $70\%$ for two models — a failure mode that cannot appear in a kinematic simulator. A second component of the discovery is that its reconstructed scenes, built from 3D Gaussian splatting with extracted collision meshes, preserve enough of the real environment that 20 paired episodes of navigation in simulation and reality give strongly correlated endpoint errors ($r = 0.935$) and mean trajectory similarity of $0.782 \pm 0.188$ nDTW. The paper takes these two results together as evidence that its benchmark measures humanoid VLN rather than a proxy for it.

Load-bearing premise

The cross-embodiment rankings rest on the simulator's dynamics matching all four robots faithfully, but the paper checks only one robot against the real world, and only on 20 episodes in two scenes.

Editorial extensions

If this is right

  • Model rankings are embodiment-dependent: the same VLN checkpoint can move from best to worst across robots, so a single canonical humanoid is insufficient for benchmarking.
  • Kinematic simulators overstate humanoid navigation performance: they cannot register the tall robot's fall rates of $64.5\%$ and $71.0\%$ or the resulting loss of success.
  • Fall rate becomes a reportable VLN metric, capturing gait stability under language-guided control alongside success and path fidelity.
  • Reconstructed scenes can substitute for artist-authored environments in physical VLN evaluation, since the pilot shows endpoint errors transfer within $0.68$ m on average.
  • The benchmark's fixed 933-episode zero-shot evaluation set allows off-the-shelf VLN checkpoints to be compared across embodiments without any training on the benchmark.

Reading between the lines

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

  • It follows, though the paper does not develop it, that the fall detector could serve as a training signal: optimizing a VLN policy jointly for instruction success and locomotion stability would likely close the gap between the most accurate model and the most stable model.
  • The $0.68$ m average endpoint gap could be used as a calibration offset to predict real-world success from simulated success for future robots, but only if the correlation replicates across more models and scenes.
  • The $100\,\mathrm{m}^2$ navigability filter implies that future humanoid VLN benchmarks should report traversable floor area per scene as standard metadata, or comparisons across legged platforms will be confounded by scene topology.
  • Scaling the pilot to multiple models and scenes is the direct test that would generalize the correlation; the paper's own stated limitation is that the current evidence is one model, two scenes, and 20 paired episodes.
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

3 major / 6 minor

Summary. The paper introduces HumanoidVLN, a physics-grounded simulator and benchmark for vision-language navigation with bipedal humanoid robots. Built on NVIDIA Isaac Sim, it provides a hierarchical control stack (a per-embodiment RL locomotion policy with interchangeable PD or MPC path trackers) and supports four embodiments: Unitree G1, Unitree H1, Internal-A, and Internal-B. The benchmark contains 87 scenes, filtered for at least 100 m^2 of navigable area, and 933 episodes with instructions produced by a multi-agent generator-reviewer-paraphraser pipeline with human-in-the-loop verification. Four VLN models (NaVILA, StreamVLN, DualVLN, JanusVLN) are evaluated zero-shot across the four embodiments, with JanusVLN reporting the highest mean SR of 43.55%. A 20-episode sim-to-real pilot with DualVLN on the Unitree G1 reports a strong correlation in endpoint navigation error (r = 0.935) and mean trajectory similarity of 0.782 nDTW. The central claim is that the platform provides physically executable, embodiment-aware evaluation that reveals meaningful cross-embodiment differences in navigation performance and fall rate.

Significance. If the physical-fidelity claim holds, HumanoidVLN addresses a genuine gap in the VLN benchmark literature: existing simulators largely rely on kinematic teleportation and do not model bipedal locomotion constraints or morphology-dependent camera dynamics. The paper has notable strengths: the evaluation is zero-shot with public checkpoints, the proposed benchmark includes human-verified instructions, the architecture is designed to be extensible to new robots and models, and the sim-to-real pilot provides a concrete, falsifiable check on the reconstructed-scene pipeline. The reported cross-embodiment fall-rate differences, if validated, would be an important new axis of evaluation for humanoid VLN. However, the current evidence supports physical fidelity for only one embodiment and only for endpoint/trajectory metrics, while the headline cross-embodiment fall rates involve robots whose simulated dynamics have not been validated against hardware. The benchmark's central contribution therefore depends on validation and analysis that is not yet present in the manuscript.

major comments (3)
  1. [Sec. IV-C] The episode sampling rule resamples any path that "cannot be completed stably," but the manuscript does not report how often resampling occurs, for which embodiments, or how resampling changes the distribution of path lengths, turning frequency, and task difficulty. Because Table III compares models across four embodiments, an embodiment-specific resampling process can confound the reported rankings: the episode sets may differ across robots in difficulty, so the observed SR and fall-rate differences may reflect sampling bias rather than morphology. The authors should report per-embodiment resampling rates and compare the final episode distributions across embodiments, and ideally construct a single common episode set verified as executable by all four robots, with a sensitivity analysis showing that the Table III rankings are stable under the choice of episode set.
  2. [Sec. VI-D and Table III] The sim-to-real pilot validates only the Unitree G1, only endpoint navigation error and trajectory nDTW, and only two scenes with one model. It does not validate the Fall Rate metric defined in Eq. (1) or any of the other three embodiments. This is a load-bearing gap because Table III reports extreme fall rates for Unitree H1 (70.95% for NaVILA, 64.52% for StreamVLN) while the other embodiments remain below 10%; if H1's simulated dynamics, its locomotion policy, or the fall-detection thresholds are even moderately inaccurate, those dramatic differences could be simulation artifacts rather than embodied facts. The conclusion's limitation paragraph lists scene diversity, human verification, and compute cost, but omits the unvalidated dynamics of three of the four embodiments. The authors should add real-robot fall-rate validation for at least one additional embodiment, or provide a systematic dynamics sensitivity analysis (e.g., friction, mass distribution, actuator gains, control latency) with confidence intervals for the Table III fall rates, and they should release the full H1 and Internal-A/B specifications currently withheld in Table II so the experiments can be reproduced.
  3. [Sec. VI-C] The 3DGS reconstruction quality assessment is purely qualitative: Fig. 4 shows normal maps, but no quantitative metrics are reported. Since reconstructed scenes form part of the benchmark, and since the collision meshes derived from 3DGS directly affect physical executability, the claim that the reconstructions are "simulation-ready" needs quantitative support. The authors should report reconstruction accuracy (for example, depth error against reference scans or mesh accuracy), and should compare physical-execution statistics such as fall rate or footstep collision rate on reconstructed versus artist-designed scenes. Without this, Q3's conclusion is not supported by the evidence presented.
minor comments (6)
  1. [Sec. V-C] The manuscript says one instruction per episode is selected using a "fixed, approximately balanced assignment" across the four styles, but it does not specify the exact assignment mechanism or release the mapping; please provide the complete procedure so that the evaluation can be reproduced exactly.
  2. [Table II] Withholding the specifications of Internal-A and Internal-B for double-blind review is understandable during reviewing, but the final version must include full specifications (or a supplement), because a benchmark episode set and cross-embodiment comparisons cannot be independently reconstructed without them.
  3. [Fig. 4] The caption states that the full pipeline "yields the most coherent geometry," but the figure is qualitative; adding a quantitative caption metric or referring to a table of reconstruction errors would make the claim verifiable.
  4. [Table III / Fig. 3] The fall rates are computed over 933 episodes, but no confidence intervals or standard errors are reported; some of the differences highlighted in the text (for example, 7.93% versus 9.54% for G1) may be within sampling noise, so interval estimates are important for interpreting the rankings.
  5. [Sec. IV-C] The sentence "paths that cannot be completed stably are resampled" should state the resampling budget and the number of rejected samples, as also requested in Major Comment 1; this will help readers assess the cost and possible bias of the procedure.
  6. [References] Please proofread the bibliography entries for consistency, as at least one entry ([24]) appears to have an inconsistent author list.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: zero-shot evaluation, independent sim-to-real measurement, and no load-bearing self-citation.

full rationale

The evaluation protocol is zero-shot with public checkpoints, so Table III results are not fitted to the benchmark or produced by construction from the benchmark's own labels. The sim-to-real pilot in Sec. VI-D compares the same DualVLN checkpoint in simulation and on the real Unitree G1, reporting a measured correlation (r=0.935) rather than a quantity that is equal to its input by definition. Episode construction does resample paths that cannot be completed stably, but this is a dataset-selection rule, not a derivation that makes any reported metric equivalent to the selection criterion; fall rates and navigation errors are still empirical outputs of the simulator. The paper references VLNVerse and other prior work for the generator-reviewer paradigm, but these are external citations and no load-bearing argument reduces to a self-citation or to a uniqueness claim imported from the authors' own prior work. The omitted specifications and lack of real-world validation for Internal-A, Internal-B, and Unitree H1 are external-validity and reproducibility limitations, not circular reasoning.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The benchmark rests on standard engineering assumptions about simulator fidelity and data curation, not on free parameters fitted to a target result. The most consequential choices are the hand-set thresholds (area, success radius, fall criteria) and the untested assumption that the simulation faithfully represents all four embodiments.

free parameters (4)
  • Navigable area threshold = 100 m^2
    Chosen by hand to exclude small scenes; directly determines the 87-scene suite composition (Sec IV-A).
  • Goal success radius = 3.0 m
    Standard VLN metric threshold; used to compute SR/OS/SPL (Sec VI-A).
  • Fall detection thresholds = T1: 0.5H_e, 1.2 m/s; T2: 0.5H_e for 2 s; T3: 0.35H_e-0.5H_e, 1.5 m/s
    Hand-set criteria for detecting falls, directly defining the reported Fall Rate metric (Eq. 1).
  • Episode resampling rule = N/A (design rule)
    Paths that cannot be completed stably are resampled (Sec IV-C); this is a hand-chosen filter that shapes the episode set.
assumptions (5)
  • domain assumption Isaac Sim rigid-body dynamics accurately model the locomotion of all four humanoid embodiments.
    Invoked throughout Sec III-B; only Unitree G1 receives a 20-episode real-world check (Sec VI-D), so H1, Internal-A, and Internal-B sim fidelity is unvalidated.
  • domain assumption Per-embodiment RL locomotion policies are stable and representative of the real robots.
    Sec III-B states the low-level RL policy without training details, rewards, or sim-to-real transfer results, so the gaits used for evaluation are assumed representative.
  • domain assumption TSDF-fused collision meshes from 3DGS reconstructions are accurate enough for physics-based navigation.
    Sec IV-B describes the pipeline; the only evidence is a qualitative ablation (Fig. 4), not a quantitative collision-accuracy assessment.
  • ad hoc to paper Resampling unstable paths does not bias the episode distribution across embodiments.
    Sec IV-C resamples paths that cannot be completed stably; if the filtering robot differs in capability from other embodiments, the benchmark may over-represent easy episodes for some robots and under-represent them for others.
  • standard math COLMAP-to-Isaac-Sim coordinate transform is correctly implemented.
    Sec IV-B applies a fixed axis-switching transform; no validation of the alignment is provided, yet all reconstructed scenes depend on it.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HumanoidVLN: A Physics-Grounded Simulator and Benchmark for Vision-Language Navigation Across Diverse Humanoid Embodiments." pith.science (2026). https://pith.science/paper/AAPPTBU2

@misc{pith2026260812860,
  author       = {Pith},
  title        = {Pith review of: HumanoidVLN: A Physics-Grounded Simulator and Benchmark for Vision-Language Navigation Across Diverse Humanoid Embodiments},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AAPPTBU2}},
  note         = {Machine review of arXiv:2608.12860}
}
read the original abstract

Vision-Language Navigation (VLN) for humanoid robots poses challenges existing benchmarks fail to address: bipedal locomotion imposes physical constraints absent from wheeled agents, humanoid morphologies vary across platforms, and egocentric observations are distorted by locomotion-induced camera dynamics. We present HumanoidVLN, a physics-grounded simulator and benchmark for VLN across diverse humanoid embodiments. Built on NVIDIA Isaac Sim, our platform supports an extensible set of humanoid configurations, demonstrated on four robots (Unitree G1, Unitree H1, Internal-A, Internal-B) spanning 10-12 lower-body DoF and heights from 1.17m to 1.80m, via a hierarchical control stack combining a reinforcement learning locomotion policy with interchangeable PD or MPC path trackers. New robots and VLN models integrate with minimal effort; we demonstrate compatibility with NaVILA, DualVLN, StreamVLN, and JanusVLN. Environments are drawn from artist-designed scenes and 3D Gaussian Splatting reconstructions, filtered for navigable areas exceeding 100 square meters. Instructions are generated by a dual generator-reviewer plus paraphraser multi-agent pipeline with human-in-the-loop verification, yielding 933 collision-aware reference episodes, each paired with one fine-grained instruction and three coarse-grained stylistic variants (formal, natural, casual). Across four models and four embodiments, JanusVLN achieves the highest mean success rate of 43.55% and nDTW of 48.38. In a 20-episode sim-to-real pilot with DualVLN and the Unitree G1, navigation errors correlate strongly (r=0.935), with a mean absolute difference of 0.68m and mean trajectory similarity of 0.782 (+/-0.188) nDTW. These results highlight the interaction between VLN models, controllers, and humanoid embodiments under physical execution. Code, benchmark, and data will be released upon acceptance at https://humanoid-vln.github.io/.

Figures

Figures reproduced from arXiv: 2608.12860 by the authors.

Figure 1
Figure 1. Overview of the HumanoidVLN physics-grounded benchmark pipeline, from embodiment-diverse humanoid platforms and hierarchical control to scene curation, episode construction, multimodal dataset generation, and plug-and￾play VLN evaluation. Abstract— Vision-Language Navigation (VLN) for humanoid robots presents unique challenges that existing benchmarks fail to address: bipedal locomotion imposes physical constraints … view at source ↗
Figure 2
Figure 2. The proposed Multi-Agent Annotation (MAA) framework. Two generator agents independently derive a route graph from the egocentric keyframe sequence alone; their graphs are reconciled, and residual contradictions are forwarded to a reviewer agent that verifies against scene-graph, trajectory, and occupancy priors. The verified instruction is paraphrased into three stylistic variants and corrected by human annotators t… view at source ↗
Figure 3
Figure 3. Fall rate (% of n = 933 episodes) across four humanoid embodiments and four VLN models. Heatmap confirming DualVLN is the most stable model across all embodiments. DualVLN leads the remaining models in nDTW (43.39), indicating stronger path adherence. NaVILA exhibits the largest average OS–SR gap (17.39 points), consistent with less reliable final stopping decisions. StreamVLN records the lowest SR (23.63%) and nDTW… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Qualitative geometry ablation of our 3DGS reconstruction pipeline. We visualize surface normals for two indoor [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Real-world validation at the episode level ( [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Path and instruction statistics for HumanoidVLN (n = 933). (a–b) Path- and instruction-length distributions. (c) Path versus instruction length across four styles (r = 0.29). [17] A. Ku, P. Anderson, R. Patel, E. Ie, and J. Baldridge, “Room￾across-room: Multilingual vi…
Figure 7
Figure 7. Figure 7: Scene statistics for HumanoidVLN. (a) Navigable area of 87 scenes (blue: artist-designed; orange: GS; dashed: 100 m2 ). (b) Counts of 17 indoor classes in six domains. scene adaptation for vision-and-language navigation,” arXiv preprint arXiv:2501.17403, 2025. [26] V. …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 22 canonical work pages

  1. [1]

    Optimization based full body control for the atlas robot,

    S. Feng, E. Whitman, X. Xinjilefu, and C. G. Atkeson, “Optimization based full body control for the atlas robot,” in2014 IEEE-RAS International Conference on Humanoid Robots, pp. 120–127, IEEE, 2014

    work page 2014

  2. [2]

    Humanoid locomotion and manipulation: Current progress and challenges in control, planning, and learning,

    Z. Gu, J. Li, W. Shen, W. Yu, Z. Xie, S. McCrory, X. Cheng, A. Shamsah, R. Griffin, C. K. Liu,et al., “Humanoid locomotion and manipulation: Current progress and challenges in control, planning, and learning,”arXiv preprint arXiv:2501.02116, 2025

    arXiv 2025

  3. [3]

    Vision-and-language nav- igation: Interpreting visually-grounded navigation instructions in real environments,

    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 nav- igation: Interpreting visually-grounded navigation instructions in real environments,” inProceedings of the IEEE conference on computer vision and pattern recognition, pp. 3674–3683, 2018

    work page 2018

  4. [4]

    Collaborative dialogue in minecraft,

    A. Narayan-Chen, P. Jayannavar, and J. Hockenmaier, “Collaborative dialogue in minecraft,” inProceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 5405–5415, 2019

    work page 2019

  5. [5]

    Matterport3d: Learning from rgb-d data in indoor environments,

    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,”arXiv preprint arXiv:1709.06158, 2017

    arXiv 2017

  6. [6]

    Grutopia: Dream general robots in a city at scale,

    H. Wang, J. Chen, W. Huang, Q. Ben, T. Wang, B. Mi, T. Huang, S. Zhao, Y . Chen, S. Yang, P. Cao, W. Yu, Z. Ye, J. Li, J. Long, Z. Wang, H. Wang, Y . Zhao, Z. Tu, Y . Qiao, D. Lin, and P. Jiangmiao, “Grutopia: Dream general robots in a city at scale,” inarXiv, 2024

    work page 2024

  7. [7]

    Towards physically executable 3d gaussian for embodied navigation,

    B. Miao, R. Wei, Z. Ge, S. Gao, J. Zhu, R. Wang, S. Tang, J. Xiao, R. Tang, J. Li,et al., “Towards physically executable 3d gaussian for embodied navigation,”arXiv preprint arXiv:2510.21307, 2025

    arXiv 2025

  8. [8]

    3d gaussian splatting for real-time radiance field rendering.,

    B. Kerbl, G. Kopanas, T. Leimk ¨uhler, G. Drettakis,et al., “3d gaussian splatting for real-time radiance field rendering.,”ACM Trans. Graph., vol. 42, no. 4, pp. 139–1, 2023

    work page 2023

Show all 40 references
  1. [9]

    Vlnverse: A benchmark for vision- language navigation with versatile, embodied, realistic simulation and evaluation,

    S. Lin, Z. Li, X. Zhao, G. Zhou, L. Wang, R. Wei, R. Tang, J. Li, H. Wang, J. Pang,et al., “Vlnverse: A benchmark for vision- language navigation with versatile, embodied, realistic simulation and evaluation,”arXiv preprint arXiv:2512.19021, 2025

    2025

  2. [10]

    Habitat: A platform for embodied ai research,

    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, pp. 9339–9347, 2019

    2019

  3. [11]

    Ai2-thor: An inter- active 3d environment for visual ai,

    E. Kolve, R. Mottaghi, W. Han, E. VanderBilt, L. Weihs, A. Herrasti, M. Deitke, K. Ehsani, D. Gordon, Y . Zhu,et al., “Ai2-thor: An inter- active 3d environment for visual ai,”arXiv preprint arXiv:1712.05474, 2017

    2017 arXiv

  4. [12]

    Airsim: High-fidelity visual and physical simulation for autonomous vehicles,

    S. Shah, D. Dey, C. Lovett, and A. Kapoor, “Airsim: High-fidelity visual and physical simulation for autonomous vehicles,” inField and service robotics: Results of the 11th international conference, pp. 621– 635, Springer, 2017

    2017

  5. [13]

    Rethinking the embodied gap in vision- and-language navigation: A holistic study of physical and visual disparities,

    L. Wang, X. Xia, H. Zhao, H. Wang, T. Wang, Y . Chen, C. Liu, Q. Chen, and J. Pang, “Rethinking the embodied gap in vision- and-language navigation: A holistic study of physical and visual disparities,” inProceedings of the IEEE/CVF International Conference on Computer Vision,...

    2025

  6. [14]

    3d gaussian ray tracing: Fast tracing of particle scenes,

    N. Moenne-Loccoz, A. Mirzaei, O. Perel, R. de Lutio, J. M. Esturo, G. State, S. Fidler, N. Sharp, and Z. Gojcic, “3d gaussian ray tracing: Fast tracing of particle scenes,”ACM Transactions on Graphics and SIGGRAPH Asia, 2024

    2024

  7. [15]

    3dgut: Enabling distorted cameras and secondary rays in gaussian splatting,

    Q. Wu, J. Martinez Esturo, A. Mirzaei, N. Moenne-Loccoz, and Z. Gojcic, “3dgut: Enabling distorted cameras and secondary rays in gaussian splatting,”Conference on Computer Vision and Pattern Recognition (CVPR), 2025

    2025

  8. [16]

    Stay on the path: Instruction fidelity in vision-and-language naviga- tion,

    V . Jain, G. Magalhaes, A. Ku, A. Vaswani, E. Ie, and J. Baldridge, “Stay on the path: Instruction fidelity in vision-and-language naviga- tion,” inProceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 1862–1872, 2019. 0 10 20 30 0 0.1 Pa...

    2019

  9. [17]

    Room- across-room: Multilingual vision-and-language navigation with dense spatiotemporal grounding,

    A. Ku, P. Anderson, R. Patel, E. Ie, and J. Baldridge, “Room- across-room: Multilingual vision-and-language navigation with dense spatiotemporal grounding,” inProceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 4392–4412, 2020

    2020

  10. [18]

    Reverie: Remote embodied visual referring expression in real indoor environments,

    Y . Qi, Q. Wu, P. Anderson, X. Wang, W. Y . Wang, C. Shen, and A. v. d. Hengel, “Reverie: Remote embodied visual referring expression in real indoor environments,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 9982–9991, 2020

    2020

  11. [19]

    Soon: Scenario oriented object navigation with graph-based exploration,

    F. Zhu, X. Liang, Y . Zhu, Q. Yu, X. Chang, and X. Liang, “Soon: Scenario oriented object navigation with graph-based exploration,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 12689–12699, 2021

    2021

  12. [20]

    Vision- and-dialog navigation,

    J. Thomason, M. Murray, M. Cakmak, and L. Zettlemoyer, “Vision- and-dialog navigation,” inConference on Robot Learning, pp. 394– 406, PMLR, 2020

    2020

  13. [21]

    Vision-based nav- igation with language-based assistance via imitation learning with indirect intervention,

    K. Nguyen, D. Dey, C. Brockett, and B. Dolan, “Vision-based nav- igation with language-based assistance via imitation learning with indirect intervention,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12527–12537, 2019

    2019

  14. [22]

    Beyond the nav-graph: Vision-and-language navigation in continuous environ- ments,

    J. Krantz, E. Wijmans, A. Majumdar, D. Batra, and S. Lee, “Beyond the nav-graph: Vision-and-language navigation in continuous environ- ments,” inEuropean Conference on Computer Vision, pp. 104–120, Springer, 2020

    2020

  15. [23]

    Towards long-horizon vision-language navigation: Platform, benchmark and method,

    X. Song, W. Chen, Y . Liu, W. Chen, G. Li, and L. Lin, “Towards long-horizon vision-language navigation: Platform, benchmark and method,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12078–12088, 2025

    2025

  16. [24]

    Human-aware vision-and-language navigation: Bridging simulation to reality with dynamic human inter- actions,

    H. Li, M. Li, Z.-Q. Cheng, Y . Dong, Y . Zhou, J.-Y . He, Q. Dai, T. Mi- tamura, and A. G. Hauptmann, “Human-aware vision-and-language navigation: Bridging simulation to reality with dynamic human inter- actions,”Advances in Neural Information Processing Systems, vol. 37, pp. ...

    2024

  17. [25]

    H. Hong, Y . Qiao, S. Wang, J. Liu, and Q. Wu, “General 1 40 87 0 1,000 2,000 Scene rank Area (m2) (a)Scene area n=87, med.=266, mean=387 HouseCafeSuper.Rest.StoreArtEx. Lib.Mus.HallOfficePantryClassConf.Show.StudioHosp.Gym 0 25 50 Count (b)Scene classes 17 classes, 6 domains ...

    2025 arXiv

  18. [26]

    gsplat: An open-source library for gaussian splatting,

    V . Ye, R. Li, J. Kerr, M. Turkulainen, B. Yi, Z. Pan, O. Seiskari, J. Ye, J. Hu, M. Tancik, and A. Kanazawa, “gsplat: An open-source library for gaussian splatting,”Journal of Machine Learning Research, vol. 26, no. 34, pp. 1–17, 2025

    2025

  19. [27]

    2d gaussian splatting for geometrically accurate radiance fields,

    B. Huang, Z. Yu, A. Chen, A. Geiger, and S. Gao, “2d gaussian splatting for geometrically accurate radiance fields,” inSIGGRAPH 2024 Conference Papers, Association for Computing Machinery, 2024

    2024

  20. [28]

    Sparse voxels ras- terization: Real-time high-fidelity radiance field rendering,

    C. Sun, J. Choe, C. Loop, W. Ma, and Y . F. Wang, “Sparse voxels ras- terization: Real-time high-fidelity radiance field rendering,” inCVPR, 2025

    2025

  21. [29]

    Habitat-gs: A high-fidelity navigation simulator with dynamic gaussian splatting,

    Z. Xia, J. Xu, C. Cui, Y . Yu, J. Zhang, Q. Yan, T. Ni, J. Chen, X. Zhou, H. Bao, R. Hu, and S. Peng, “Habitat-gs: A high-fidelity navigation simulator with dynamic gaussian splatting,” 2026

    2026

  22. [30]

    Pgsr: Planar-based gaussian splatting for efficient and high-fidelity surface reconstruction,

    D. Chen, H. Li, W. Ye, Y . Wang, W. Xie, S. Zhai, N. Wang, H. Liu, H. Bao, and G. Zhang, “Pgsr: Planar-based gaussian splatting for efficient and high-fidelity surface reconstruction,”IEEE Transactions on Visualization and Computer Graphics, vol. 31, no. 9, p. 6100–6111, 2025

    2025

  23. [31]

    Kinectfusion: Real-time dense surface mapping and tracking,

    R. A. Newcombe, S. Izadi, O. Hilliges, D. Molyneaux, D. Kim, A. J. Davison, P. Kohi, J. Shotton, S. Hodges, and A. Fitzgibbon, “Kinectfusion: Real-time dense surface mapping and tracking,” in 2011 10th IEEE International Symposium on Mixed and Augmented Reality, pp. 127–136, 2011

    2011

  24. [32]

    Structure-from-motion revisited,

    J. L. Sch ¨onberger and J.-M. Frahm, “Structure-from-motion revisited,” inConference on Computer Vision and Pattern Recognition (CVPR), 2016

    2016

  25. [33]

    Pixel- wise view selection for unstructured multi-view stereo,

    J. L. Sch ¨onberger, E. Zheng, M. Pollefeys, and J.-M. Frahm, “Pixel- wise view selection for unstructured multi-view stereo,” inEuropean Conference on Computer Vision (ECCV), 2016

    2016

  26. [34]

    Qwen3-vl technical report,

    S. Bai, Y . Cai, R. Chen, K. Chen, X. Chen, Z. Cheng, L. Deng, W. Ding, C. Gao, C. Ge,et al., “Qwen3-vl technical report,”arXiv preprint arXiv:2511.21631, 2025

    2025 arXiv

  27. [35]

    Gemma 4 technical report,

    G. Team, “Gemma 4 technical report,” 2026

    2026

  28. [36]

    Internvl3. 5: Advancing open-source multi- modal models in versatility, reasoning, and efficiency,

    W. Wang, Z. Gao, L. Gu, H. Pu, L. Cui, X. Wei, Z. Liu, L. Jing, S. Ye, J. Shao,et al., “Internvl3. 5: Advancing open-source multi- modal models in versatility, reasoning, and efficiency,”arXiv preprint arXiv:2508.18265, 2025

    2025 arXiv

  29. [37]

    Navila: Legged robot vision-language- action model for navigation,

    A.-C. Cheng, Y . Ji, Z. Yang, Z. Gongye, X. Zou, J. Kautz, E. Bıyık, H. Yin, S. Liu, and X. Wang, “Navila: Legged robot vision-language- action model for navigation,”arXiv preprint arXiv:2412.04453, 2024

    2024 arXiv

  30. [38]

    Streamvln: Streaming vision-and- language navigation via slowfast context modeling,

    M. Wei, C. Wan, X. Yu, T. Wang, Y . Yang, X. Mao, C. Zhu, W. Cai, H. Wang, Y . Chen,et al., “Streamvln: Streaming vision-and- language navigation via slowfast context modeling,”arXiv preprint arXiv:2507.05240, 2025

    2025 arXiv

  31. [39]

    Ground slow, move fast: A dual-system foundation model for generalizable vision-and-language navigation,

    M. Wei, C. Wan, J. Peng, X. Yu, Y . Yang, D. Feng, W. Cai, C. Zhu, T. Wang, J. Pang,et al., “Ground slow, move fast: A dual-system foundation model for generalizable vision-and-language navigation,” arXiv preprint arXiv:2512.08186, 2025

    2025

  32. [40]

    Janusvln: Decoupling semantics and spatiality with dual implicit memory for vision-language navigation,

    S. Zeng, D. Qi, X. Chang, F. Xiong, S. Xie, X. Wu, S. Liang, M. Xu, X. Wei, and N. Guo, “Janusvln: Decoupling semantics and spatiality with dual implicit memory for vision-language navigation,”arXiv preprint arXiv:2509.22548, 2025

    2025

Pith tools

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