RNNs can generate bounded hierarchical languages with optimal memory
Comments EMNLP2020 + appendix typo fixes
作者
Percy Liang
Natural Language Processing
The EOS Decision and Length Extrapolation
Comments 16 page, 7 Figures, 9 Tables, Blackbox NLP Workshop at EMNLP 2020
Learning Adaptive Language Interfaces through Decomposition
Comments Accepted at the 1st Workshop for Interactive and Executable Semantic Parsing (IntEx-SemPar) @ EMNLP 2020. 11 pages, 5 figures
On the Importance of Adaptive Data Collection for Extremely Imbalanced Pairwise Tasks
Comments In Findings of EMNLP 2020
Enabling Language Models to Fill in the Blanks
Comments Published as a conference paper at ACL 2020
An Investigation of Why Overparameterization Exacerbates Spurious Correlations
Robustness to Spurious Correlations via Human Annotations
Comments ICML 2020 final version, 16 pages, 9 figures
Learning Abstract Models for Strategic Exploration and Fast Reward Transfer
Understanding and Mitigating the Tradeoff Between Robustness and Accuracy
Comments Appearing at International Conference on Machine Learning (ICML) 2020
Graph-based, Self-Supervised Program Repair from Diagnostic Feedback
Comments ICML 2020. Code & data available at https://github.com/michiyasunaga/DrRepair
Selective Question Answering under Domain Shift
Comments ACL 2020
Shaping Visual Representations with Language for Few-shot Classification
Comments ACL 2020. Version 1 appeared at the NeurIPS 2019 Workshop on Visually Grounded Interaction and Language (ViGIL)
ExpBERT: Representation Engineering with Natural Language Explanations
Comments ACL 2020
Robust Encodings: A Framework for Combating Adversarial Typos
Comments ACL 2020
Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization
Understanding Self-Training for Gradual Domain Adaptation
Strategies for Pre-training Graph Neural Networks
Comments Accepted as a spotlight to ICLR 2020
Verified Uncertainty Calibration
Comments Accepted as a spotlight to NeurIPS 2019, updated to include experiments for ECE
On the Accuracy of Influence Functions for Measuring Group Effects
Learning Autocomplete Systems as a Communication Game
A Tight Analysis of Greedy Yields Subexponential Time Approximation for Uniform Decision Tree
Comments 40 pages, 5 figures
Designing and Interpreting Probes with Control Tasks
Comments EMNLP 2019
Distributionally Robust Language Modeling
Comments Camera ready version for EMNLP
Certified Robustness to Adversarial Word Substitutions
Comments EMNLP 2019
Adversarial Training Can Hurt Generalization
SPoC: Search-based Pseudocode to Code
Comments Under submission to NeurIPS 2019
Maximum Weighted Loss Discrepancy
Comments ICLR 2019 Workshop. Safe Machine Learning: Specification, Robustness, and Assurance
Pun Generation with Surprise
Comments NAACL 2019
Unifying Human and Statistical Evaluation for Natural Language Generation
Comments NAACL Camera Ready Submission
Defending against Whitebox Adversarial Attacks via Randomized Discretization
Comments In proceedings of the 22nd International Conference on Artificial Intelligence and Statistics
