collaborators

6 papers

cs.RO2026

Good in Bad (GiB): Sifting Through End-user Demonstrations for Learning a Better Policy

Noushad Sojib, Ola Ghattas, Momotaz Begum

Imitation learning offers a promising framework for enabling robots to acquire diverse skills from human users. However, most imitation learning algorithms assume access to high-qu…

cs.RO2026

Trajectory-Consistent Flow Matching for Robust Visuomotor Policy Learning

Riad Ahmed, Sujosh Nag, Moniruzzaman Akash +2

Flow matching policies learn continuous velocity fields that transport noise to actions, enabling fast deterministic inference for robot manipulation. However, standard training op…

cs.RO2026

TAIL-Safe: Task-Agnostic Safety Monitoring for Imitation Learning Policies

Riad Ahmed, Momotaz Begum

Recent imitation learning (IL) algorithms such as flow-matching and diffusion policies demonstrate remarkable performance in learning complex manipulation tasks. However, these pol…

cs.RO2026

An Efficient Metric for Data Quality Measurement in Imitation Learning

Noushad Sojib, Momotaz Begum

Imitation learning (IL) has seen remarkable progress, yet field deployment of IL-powered robots remains hindered by the challenge of out-of-distribution (OOD) scenarios. Fine-tunin…

cs.RO2026

A Principled Approach for Creating High-fidelity Synthetic Demonstrations for Imitation Learning

Moniruzzaman Akash, Momotaz Begum

Recent advances in 3D Gaussian Splatting (3DGS) have enabled visually realistic demonstration generation from a single expert trajectory and a short multi-view scan. However, exist…

cs.RO2026

To Do or Not to Do: Ensuring the Safety of Visuomotor Policies Learned from Demonstrations

Riad Ahmed, Moniruzzaman Akash, Momotaz Begum

Task success has historically been the primary measure of policy performance in imitation learning (IL) research. This characteristics strictly limits the ubiquitous applications o…