Ireland time = GMT+1

GMT+1 GMT EST

08:50–09:00 07:50–08:00 03:50–04:00 Opening remark

09:00–09:45 08:00–08:45 04:00–04:45 Invited talk: Sebastian Riedel Patrick Lewis

09:45–10:30 08:45–09:30 04:45–05:30 Invited talk: Monojit Choudhury

10:30–10:45 09:30–09:45 05:30–05:45 Oral Papers Spotlight Presentations

10:45–11:00 09:45–10:00 05:45–06:00 Coffee break

11:00–12:00 10:00–11:00 06:00–07:00 Poster session I (Virtual + in-person)

12:00–13:30 11:00–12:30 07:00–08:30 Lunch break

13:30–14:15 12:30–13:15 08:30–09:15 Invited talk: Emma Strubell

14:15–15:15 13:15–14:15 09:15–10:15 Poster session II (Virtual + in-person)

15:15–15:30 14:15–14:30 10:15–10:30 Coffee break

15:30–16:15 14:30–15:15 10:30–11:15 Invited talk: Been Kim

16:15–17:00 15:15–16:00 11:15–12:00 Panel discussion

17:00–17:45 16:00–16:45 12:00–12:45 Invited talk: Percy Liang

17:45–17:50 16:45–16:50 12:45–12:50 Closing remark

Poster Assignment

These are assignment for virtual poster session. Authors attending the workshop in-person will have an in-person poster presentation in the other poster session (e.g., if the paper is assigned to Session 1 for the virtual poster, they will present in-person in the Session 2).

Poster session 1 (11:00–12:00 Irish Time)

Identifying the Limits of Cross-Domain Knowledge Transfer for Pretrained Models

Analyzing Gender Representation in Multilingual Models

ANNA: Enhanced Language Representation for Question Answering

Binary Encoded Word Mover's Distance

Clozer: Adaptable Data Augmentation for Cloze-style Reading Comprehension

Q-Learning Scheduler for Multi Task Learning Through the use of Histogram of Task Uncertainty

TRAttack: Text Rewriting Attack Against Text Retrieval

From Hyperbolic Geometry Back to Word Embeddings

On Target Representation in Continuous-output Neural Machine Translation

Detecting Textual Adversarial Examples Based on Distributional Characteristics of Data Representations

A Comparative Study of Pre-trained Encoders for Low-Resource Named Entity Recognition

Towards Improving Selective Prediction Ability of NLP Systems

Same Author or Just Same Topic? Towards Topic-Independent Style Representations

A Vocabulary-Free Multilingual Neural Tokenizer for End-to-End Task Learning

PALBERT: Teaching ALBERT to Ponder

Poster session 2 (14:15–15:15 Irish Time)

Zero-shot Cross-lingual Transfer is Under-specified Optimization

Detecting Word-Level Adversarial Text Attacks via SHapley Additive exPlanations

Isomorphic Cross-Lingual Word Embeddings for Low-Resource Languages

Unsupervised Geometric and Topological Approaches for Cross-Lingual Sentence Representation and Comparison

On the Geometry of Concreteness

Temporal Knowledge Graph Reasoning with Low-rank and Model-agnostic Representations

PARADISE: Exploiting Parallel Data for Multilingual Sequence-to-Sequence Pretraining

Distributionally Robust Recurrent Decoders with Random Network Distillation

Video Language Co-Attention with Multimodal Fast-Learning Feature Fusion for VideoQA

WeaNF: Weak Supervision with Normalizing Flows

When does CLIP generalize better than unimodal models? When judging human-centric concepts

A Study on Entity Linking Across Domains: Which Data is Best for Fine-Tuning?

Lacking the Embedding of a Word? Look it up into a Traditional Dictionary