• Computer Science
  • Published in ArXiv 2019

ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks

@article{Shridhar2019ALFREDAB,
  title={ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks},
  author={Mohit Shridhar and Jesse Thomason and Daniel Gordon and Yonatan Bisk and Winson Han and Roozbeh Mottaghi and Luke Zettlemoyer and Dieter Fox},
  journal={ArXiv},
  year={2019},
  volume={abs/1912.01734}
}
We present ALFRED (Action Learning From Realistic Environments and Directives), a benchmark for learning a mapping from natural language instructions and egocentric vision to sequences of actions for household tasks. Long composition rollouts with non-reversible state changes are among the phenomena we include to shrink the gap between research benchmarks and real-world applications. ALFRED consists of expert demonstrations in interactive visual environments for 25k natural language directives… CONTINUE READING

References

Publications referenced by this paper.
SHOWING 1-10 OF 56 REFERENCES

Self-Monitoring Navigation Agent via Auxiliary Progress Estimation

VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

Vision-and-Language Navigation: Interpreting Visually-Grounded Navigation Instructions in Real Environments

VIEW 10 EXCERPTS
HIGHLY INFLUENTIAL

TOUCHDOWN: Natural Language Navigation and Spatial Reasoning in Visual Street Environments

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

VirtualHome: Simulating Household Activities Via Programs

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

Embodied Question Answering in Photorealistic Environments With Point Cloud Perception

VIEW 2 EXCERPTS

Multi-Target Embodied Question Answering

VIEW 1 EXCERPT