July 2, 2020 | Written by: Avi Sil and Karthik Sankaranarayanan
Share this post:
The 58th Annual Meeting of the Association for Computational Linguistics (ACL 2020), the premiere annual conference on AI and language, takes place July 5-10. As is the case with most events currently, ACL will be virtual this year due to COVID-19. At IBM Research AI, we’re excited to share with you — wherever you might be in the world — all the work we’ll have at ACL 2020 designed to advance AI for the enterprise.
The ability of AI to master language has been one of IBM Research AI’s key areas of focus for years. The field of Natural Language Processing (NLP) is constantly evolving in efforts to better outfit AI with the ability to communicate similarly to how us humans can. It’s an incredibly challenging area of research. An AI must identify, decipher, and navigate through natural language barriers — tasks like slang, idioms, acronyms, different languages and extracting meaning from multi-format documents, to name a few.
To tackle these challenges, IBM released earlier this year a new, four-part mastering language taxonomy that we believe will advance enterprise AI both through enhancing basic NLP features, as well as introducing more advanced tools and concepts. To learn more about this strategy and how we’re advancing it, click here.
Technical efforts at ACL 2020
We’re looking forward to once again having a robust presence at ACL that showcases a variety of IBM Research AI’s latest efforts in AI and language, including:
- Automatic question generation (QG) for fast domain adaptation of enterprise QA systems
- Automatic argument summarization leveraging technology from IBM Project Debater
- Frequently Asked Questions (FAQ) retrieval
- Automatic taxonomy construction
- Insight analysis of the BERT (Bidirectional Encoder Representations from Transformers) NLP technique
Sponsorship and virtual booth
IBM Research AI is proudly sponsoring ACL 2020. We hope you will join us at our virtual booth to see and hear about our latest technology demos, publications and career opportunities including the AI Residency Program.
For a full list of our papers and demos at the conference, see below.
- Go Ahead Ask Me Anything (GAMMA): GAAMA is a (multi-lingual) reading comprehension system for question-answering.
- ExBERT: A Visual Tool to Explore BERT: Learn how to uncover insights into what deep Transformer models understand about human language by interactively exploring their learned attentions and contextual embeddings.
- NLQ over BAI: Natural Language Querying technology to dynamically generate insights from Business Process automation data.
- Label Noise in Context (LNiC) : Uses training set context to increase precision and add explainability to a label noise detection system. Demo tool here and video here.
- Vendor Master Scrubber: Identification and removal of duplicate vendors from vendor database using semantic clustering.
- Cognitive Invoice Processing (ADAPT) : An automated tool consisting of semantic matching and semantic classifier to assist invoice processing in account payable process.
Accepted Papers at ACL
On the Importance of Diversity in Question Generation for QA
Arafat Sultan, Shubham Chandel, Ramon Astudillo, Vittorio Castelli
Out of the Echo Chamber: Detecting Countering Debate Speeches
Matan Orbach, Yonatan Bilu, Assaf Toledo, Dan Lahav, Michal Jacovi, Ranit Aharonov, Noam Slonim
Learning Implicit Text Generation via Feature Matching
Inkit Padhi, Pierre Dognin, Ke Bai, Cicero Nogueira dos Santos, Vijil Chenthamarakshan, Youssef Mroueh, Payel Das
A Multi-Perspective Architecture for Semantic Code Search
Rajarshi Haldar, Lingfei Wu, Jinjun Xiong, Julia Hockenmaier
From Arguments to Key Points: Towards Automatic Argument Summarization
Roy Bar-haim, Lilach Edelstein, Roni Friedman-melamed, Yoav Kantor, Dan Lahav, Noam Slonim
Improving Segmentation for Technical Support Problems
Abhirut Gupta, Kushal Chauhan
Interactive Construction of User-Centric Dictionary for Text Analytics
Ryosuke Kohita, Issei Yoshida, Hiroshi Kanayama, Tetsuya Nasukawa
Unsupervised FAQ Retrieval with Question Generation and BERT
Yosi Mass, Boaz Carmeli, Haggai Roitman, David Konopnicki
Knowledge Graph-Augmented Abstractive Summarization with Semantic-Driven Cloze Reward
Luyang Huang, Lingfei Wu and Lu Wang
Crossing Variational Auto-encoders for Answer Retrieval
Wenhao Yu, Lingfei Wu, Qingkai Zeng, Yu Deng, Shu Tao, Meng Jiang
A Joint End-to-End Neural Model for Information Extraction with Global Features
Ying Lin, Heng Ji, Fei Huang, Lingfei Wu
The TechQA Dataset
Vittorio Castelli, Rishav Chakravarti, Saswati Dana, Anthony Ferritto, Hans Florian, Martin Franz, Dinesh Garg, Dinesh Khandelwal, Scott Mccarley, Mike Mccawley, Mohamed Nasr, Lin Pan, Cezar Pendus, John Pitrelli, Saurabh Pujar, Salim Roukos, Andy Sakrajda, Avi Sil, Rosario Uceda-sosa, Todd Ward, Rong Zhang
GPT-too: A language-model-first approach for AMR-to-text generation
Manuel Mager, Ramon Astudillo, Tahira Naseem, Arafat Sultan, Young-suk Lee, Hans Florian, Salim Roukos
Span Selection Pre-training for Question Answering
Michael Glass, Alfio Gliozzo, Rishav Chakravarti, Anthony Ferritto, G P Shrivatsa Bhargav, Dinesh Garg, Avi Sil
Taxonomy Construction via Graph-based Cross-Domain Knowledge Transfer
Chao Shang, Sarthak Dash, Md Faisal Mahbub Chowdhury, Nandana Mihindukulasooriya, Alfio Gliozzo,
ExBERT: A Visual Analysis Tool to Explore Learned Representations in Transformer Models
Benjamin Hoover, Hendrik Strobelt, Sebastian Gehrmann
Implicit Discourse Relation Classification: We Need to Talk About Evaluation
Najoung Kim, Song Feng, Chulaka Gunasekara, Luis Lastras
HAT: Hardware-Aware Transformers for Efficient Neural Machine Translation
Hanrui Wang, Zhanghao Wu, Zhijian Liu, Han Cai, Ligeng Zhu, Chuang Gan, Song Han
Bridging Anaphora Resolution as Question Answering
Label Noise in Context
Michael Desmond, Catherine Finegan-Dollak, Jeff Boston, Matthew Arnold