Publications

(2024). Language Models can Infer Action Semantics for Classical Planners from Environment Feedback. arXiv, 2024.

PDF

(2024). Efficient End-to-End Visual Document Understanding with Rationale Distillation. NAACL, 2024.

PDF Project Poster Slides

(2023). Chain-of-Questions Training with Latent Answers for Robust Multistep Question Answering. EMNLP, 2023.

PDF Code Poster

(2023). VLN Pretraining Still Works with Nonsensical or Irrelevant Instructions. O-DRUM@CVPR, 2023.

PDF

(2023). Iterative Vision-and-Language Navigation. CVPR, 2023.

PDF Code Project

(2022). Generalization Differences between End-to-End and Neuro-Symbolic Vision-Language Reasoning Systems. Findings of EMNLP, 2022.

PDF Code Poster Slides Video

(2022). Navigating Memory Construction by Global Pseudo-Task Simulation for Continual Learning. NeurIPS, 2022.

PDF Poster Slides Video

(2021). Learning to Generalize Compositionally by Transferring Across Semantic Parsing Tasks. arXiv, 2021.

PDF Code

(2020). BabyWalk: Going Farther in Vision-and-Language Navigation by Taking Baby Steps. ACL, 2020.

PDF Code Slides Video

(2018). Interpreting Deep Sports Analytics: Valuing Actions and Players in the NHL. MLSA@ECML-PKDD, 2018.

PDF Slides

(2018). Toward Interpretable Deep ReinforcementLearning with Linear Model U-Trees. ECML-PKDD, 2018.

PDF Code Poster