INLG2021

Home Calls Workshops Tutorial GenChal 2021 Organisation Registration Sponsorship Contact ☰

Deadline for Submissions
15 June 2021
23:59 AOE (UTC-12)

Notification
~~26 July 2021~~

Camera Ready
~~15 August 2021~~

INLG 2021
20 - 24 Sept 2021

INLG Schedule

This is the preliminary schedule, subject to change.

Time zone: UK Time (GMT+1)

INLG Schedule: Overview

Monday, 20 Sep

1200-1730: BigScience Workshop: Large Language Models (see Workshops page)

1400-1600: Tutorial on Crowdsourcing Experiments and Platforms by David M. Howcroft

Tuesday, 21 Sep

1200-1300: Welcome

Invited talk by Natalie Schluter

1320-1440: Oral session 1: Models and Questions

1440-1540: Mid-afternoon break

1540-1640: Poster session 1

1700-1800: Panel: What users want from real-world NLG

1800-1900: Socializing

Wednesday, 22 Sep

1200-1320: Oral session 2: NLG tasks, evaluation, and explanation

1340-1440: Poster session 2

1440-1540: Mid-afternoon break

1540-1620: Generation Challenges: results from existing challenges

1630-1710: Generation Challenges: Submissions to existing challenges

1720-1800: Generation Challenges: new challenges

1800-1900: SIGGEN business meeting and socializing

Thursday, 23 Sep

1200-1300: Oral session 3: Emotions and user adaptation

1320-1440: Birds of a Feather (discussions in breakout rooms)

1440-1540: Mid-afternoon break

1540-1640: Poster session 3

1700-1800: Invited talk by Tim Bickmore

Closing session

Session Details

Tuesday, 21 Sep

Oral session 1: Models and Questions (1320-1440)

Exploring Input Representation Granularity for Generating Questions Satisfying Question-Answer Congruence

Generating Diverse Descriptions from Semantic Graphs

Neural Methodius Revisited: Do Discourse Relations Help with Pre-Trained Models Too?

Towards Zero-Shot Multilingual Synthetic Question and Answer Generation for Cross-Lingual Reading Comprehension

Poster session 1 (1540-1640)

Chefbot: A Novel Framework for the Generation of Commonsense-enhanced Responses for Task-based Dialogue Systems (demo)

Examining Covert Gender Bias: A Case Study in Turkish and English Machine Translation Models

Generating Racing Game Commentary from Vision, Language, and Structured Data

Multi-Sentence Knowledge Selection in Open-Domain Dialogue

Predicting Antonyms in Context using BERT

Self-Training for Compositional Neural NLG in Task-Oriented Dialogue

WeaSuL: Weakly Supervised Dialogue Policy Learning: Reward Estimation for Multi-turn Dialogue

Wednesday, 22 Sep

Oral session 2: NLG tasks, evaluation, and explanation (1200-1320)

Explaining Decision-Tree Predictions by Addressing Potential Conflicts between Predictions and Plausible Expectations

Formulating Neural Sentence Ordering as the Asymmetric Traveling Salesman Problem

Underreporting of errors in NLG output, and what to do about it

What can Neural Referential Form Selectors Learn

Poster session 2 (1340-1440)

Contextualizing Variation in Text Style Transfer Datasets

Enriching the E2E dataset

Goal-Oriented Script Construction

HI-CMLM: Improve CMLM with Hybrid Decoder Input

SAPPHIRE: Approaches for Enhanced Concept-to-Text Generation

Single Example Can Improve Zero-Shot Data Generation

Using BERT for choosing classifiers in Mandarin

Generation Challenges: results from existing challenges (oral) (1540-1620)

Generation Challenges: Results of the Accuracy Evaluation Shared Task

The ReproGen Shared Task on Reproducibility of Human Evaluations in NLG: Overview and Results

Generation Challenges: Submissions to existing challenges (posters) (1630-1710)

Automatic Verification of Data Summaries

Grounding NBA Matchup Summaries

Shared Task in Evaluating Accuracy: Leveraging Pre-Annotations in the Validation Process

Text-in-Context: Token-Level Error Detection for Table-to-Text Generation

A Reproduction Study of an Annotation-based Human Evaluation of MT Outputs

Reproducing a comparison of hedged and non-hedged NLG texts

Another PASS - A Reproduction Study of the Human Evaluation of a Football Report Generation System

TUDA-Reproducibility @ ReproGen: Replicability of Human Evaluation of Text-To-Text and Concept-To-Text Generation

Generation Challenges: new challenges (short oral papers) (1720-1800)

DialogSum Challenge: Summarizing Real-Life Scenario Dialogues

Quality Evaluation of the Low-Resource Synthetically Generated Code-Mixed Hinglish Text

Shared Task on Feedback Comment Generation for Language Learners

The SelectGen Challenge: Finding the Best Training Samples for Few-Shot Neural Text Generation

Thursday, 23 Sep

Oral Session 3: Emotions and user adaptation (1200-1300)

Affective Decoding for Empathetic Response Generation

Controllable Sentence Simplification with a Unified Text-to-Text Transfer Transformer

SEPRG: Sentiment aware Emotion controlled Personalized Response Generation

Birds of a Feather: Discussion rooms (1320-1440)

Evaluation

Explainability and Trustworthiness

Neural NLG

NLG in Healthcare

NLG Resources (including low-resource languages)

Rule-Based NLG

Poster session 3 (1540-1640)

Attention Is Indeed All You Need: Semantically Attention-Guided Decoding for Data-to-Text NLG

BERT-based distractor generation for Swedish reading comprehension questions using a small-scale dataset

Biomedical Data-to-Text Generation via Fine-Tuning Transformers

Decoding, Fast and Slow: A Case Study on Balancing Trade-Offs in Incremental, Character-level Pragmatic Reasoning

Exploring Structural Encoding for Data-to-Text Generation

GraphPlan: Story Generation by Planning with Event Graph

Visit the INLG 2021 Registration Webpage to Register Now!

For media inquiries or questions/comments, please contact inlgmeeting@gmail.com.
For announcements, follow @inlgmeeting on Twitter and join our mailing list at http://www.jiscmail.ac.uk/SIGGEN.