TY  - RPRT
TI  - FOSSIL: Harnessing Feedback on Suboptimal Samples for Data-Efficient Generalisation with Imitation Learning for Embodied Vision-and-Language Tasks
AU  - Sabrina McCallum
AU  - Amit Parekh
AU  - Alessandro Suglia
PY  - 2025
UR  - https://arxiv.org/abs/2510.11307
ID  - 2510.11307
ER  -