Academic Publications: IEEE Xplore
Academic Publications · IEEE Xplore. IEEE journals, conferences and technical publications. This source has 4 entries admitted under this category's review policy; each links directly to the original.
Authority tiers in this category: A2–R.
NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration
Evidence grade: for discovery only; must not by itself support a formal conclusion.
The paper unifies goal-directed navigation and exploration of unknown environments with a goal-masked diffusion policy. Its ICRA 2024 Best Conference Paper award is verified, but the official award page does not explain why it is a major contribution to AI, so it is listed only as a public entry awaiting further evidence.
Evidence use: For checking the paper's identity, DOI, ICRA 2024 Best Conference Paper result and original content; it cannot currently serve as evidence of a major AI contribution confirmed by this site.
Limitations: No external statement of a major AI contribution; being listed publicly records a gap in the review and does not mean the six-step importance review was passed. The official award shows recognition under that award process, but the award page gives no committee statement on major impact on AI. Capability and performance claims in the paper are as reported by the authors; this site has not reproduced the experiments independently.
Related concepts: Diffusion Models · Reinforcement Learning · Planning and Task Decomposition
Link last checked: 2026-09-24
Open X-Embodiment: Robotic Learning Datasets and RT-X Models
Evidence grade: for discovery only; must not by itself support a formal conclusion.
The paper pools robot datasets from many institutions and studies training RT-X models across multiple robot embodiments. Its ICRA 2024 Best Conference Paper award is verified, but the available official award materials do not establish a major contribution to AI, so it is listed only as a public entry awaiting further evidence.
Evidence use: For checking the paper's identity, DOI, collaborating team, ICRA 2024 Best Conference Paper result and original content; it cannot currently serve as evidence of a major AI contribution confirmed by this site.
Limitations: No external statement of a major AI contribution; being listed publicly records a gap in the review and does not mean the six-step importance review was passed. The official award shows recognition under that award process, but the award page does not explain whether it substantially changed AI or robot learning. Claims about scale, transfer ability and performance in the paper are as reported by the authors; this site has not reproduced the experiments independently.
Related concepts: Multimodal Models · Reinforcement Learning · World Models and 3D Generation
Link last checked: 2026-09-24
Marginalizing and Conditioning Gaussians onto Linear Approximations of Smooth Manifolds with Applications in Robotics
Evidence grade: for discovery only; must not by itself support a formal conclusion.
The paper gives closed-form expressions for marginalizing and conditioning Gaussians onto linear approximations of manifolds, used for uncertainty estimation in constrained robot optimization. The ICRA 2025 committee described its robotics contribution but did not establish that it significantly changed AI, so it is listed only as a public entry awaiting further evidence.
Evidence use: For checking the paper's identity, DOI, ICRA 2025 Best Conference Paper result and the committee's statement on its robotics contribution; it cannot currently serve as evidence of a major AI contribution confirmed by this site.
Limitations: No external statement of a major AI contribution; being listed publicly records a gap in the review and does not mean the six-step importance review was passed. The committee statement supports its outstanding contribution to constrained optimization in robotics but does not state that it significantly changed the course of AI. The mathematical properties and experimental results in the paper are as reported by the authors; this site has not reproduced them independently.
Related concepts: Uncertainty Calibration and Selective Prediction · Planning and Task Decomposition
Link last checked: 2026-09-24
MAC-VO: Metrics-Aware Covariance for Learning-Based Stereo Visual Odometry
Evidence grade: for discovery only; must not by itself support a formal conclusion.
The paper proposes a learned, metrically aware covariance model for keypoint selection and residual weighting in stereo visual odometry. The ICRA 2025 committee described its contribution to robustness and accuracy, but available external materials do not show a major contribution to AI, so it is listed only as a public entry awaiting further evidence.
Evidence use: For checking the paper's identity, DOI, ICRA 2025 Best Conference Paper result and the committee's statement of contribution; it cannot currently serve as evidence of a major AI contribution confirmed by this site.
Limitations: No external statement of a major AI contribution; being listed publicly records a gap in the review and does not mean the six-step importance review was passed. The committee confirms it improves the robustness and accuracy of visual odometry but does not state that it significantly changed the course of AI. Performance and scope in the paper are as reported by the authors; this site has not reproduced the experiments independently.
Related concepts: Uncertainty Calibration and Selective Prediction · Self-supervised Learning
Link last checked: 2026-09-24