Chirag Rastogi

Marc Canby, Adam Davies, Chirag Rastogi et al.

Causal probing aims to analyze foundation models by examining how intervening on their representation of various latent properties impacts their outputs. Recent works have cast doubt on the theoretical basis of several leading causal probing methods, but it has been unclear how to systematically evaluate the effectiveness of these methods in practice. To address this, we define two key causal probing desiderata: completeness (how thoroughly the representation of the target property has been transformed) and selectivity (how little non-targeted properties have been impacted). We find that there is an inherent tradeoff between the two, which we define as reliability, their harmonic mean. We introduce an empirical analysis framework to measure and evaluate these quantities, allowing us to make the first direct comparisons between different families of leading causal probing methods (e.g., linear vs. nonlinear, or concept removal vs. counterfactual interventions). We find that: (1) all methods show a clear tradeoff between completeness and selectivity; (2) more complete and reliable methods have a greater impact on LLM behavior; and (3) nonlinear interventions are almost always more reliable than linear interventions. Our project webpage is available at: https://ahdavies6.github.io/causal_probing_reliability/

Yu Chen, Chirag Rastogi, Zheyu Zhou et al.

Determining the drivable area, or free space segmentation, is critical for mobile robots to navigate indoor environments safely. However, the lack of coherent markings and structures (e.g., lanes, curbs, etc.) in indoor spaces places the burden of traversability estimation heavily on the mobile robot. This paper explores the use of a self-supervised one-shot texture segmentation framework and an RGB-D camera to achieve robust drivable area segmentation. With a fast inference speed and compact size, the developed model, MOSTS is ideal for real-time robot navigation and various embedded applications. A benchmark study was conducted to compare MOSTS's performance with existing one-shot texture segmentation models to evaluate its performance. Additionally, a validation dataset was built to assess MOSTS's ability to perform texture segmentation in the wild, where it effectively identified small low-lying objects that were previously undetectable by depth measurements. Further, the study also compared MOSTS's performance with two State-Of-The-Art (SOTA) indoor semantic segmentation models, both quantitatively and qualitatively. The results showed that MOSTS offers comparable accuracy with up to eight times faster inference speed in indoor drivable area segmentation.

Sharan Balasubramanian, Ayush Rajput, Rodra W. Hascaryo et al.

Autonomous Underwater Vehicles (AUVs) and Remotely Operated Vehicles (ROVs) are used for a wide variety of missions related to exploration and scientific research. Successful navigation by these systems requires a good localization system. Kalman filter based localization techniques have been prevalent since the early 1960s and extensive research has been carried out using them, both in development and in design. It has been found that the use of a dynamic model (instead of a kinematic model) in the Kalman filter can lead to more accurate predictions, as the dynamic model takes the forces acting on the AUV into account. Presented in this paper is a motion-predictive extended Kalman filter (EKF) for AUVs using a simplified dynamic model. The dynamic model is derived first and then it was simplified for a RexROV, a type of submarine vehicle used in simple underwater exploration, inspection of subsea structures, pipelines and shipwrecks. The filter was implemented with a simulated vehicle in an open-source marine vehicle simulator called UUV Simulator and the results were compared with the ground truth. The results show good prediction accuracy for the dynamic filter, though improvements are needed before the EKF can be used on real-time. Some perspective and discussion on practical implementation is presented to show the next steps needed for this concept.