Notes from SIGIR 2022
KeyNotes:
From Search Queries to Conversations in the Design of Information Retrieval and Access Systems- I3A: An Intelligent Interactive Information Agent Model for Information Retrieva
- Few-shot Information Extraction is here: Pre-train, Prompt and Entail
- Intelligent Conversational Agents for Ambient Computing
- Search at Bloomberg: Challenges, Opportunities and Lessons Learned
- Amazon: The Role Of GNNs In Improving Search Quality And Trust
- Panel: Applications and Future of Dense Retrieval in Industry
Notable Papers:
Structure and Semantics Preserving Document Representations
Paper: 
|
Recording:
|
BERT Rankers are Brittle: a Study using Adversarial Document Perturbations
|
Paper:
|
Recording:
|
An Industrial Framework for Cold-Start Recommendation in Zero-Shot Scenarios
Note: RecSys at AntGroup scale…
|
Paper:
|
Recording:
|
A Non-Factoid Question-Answering Taxonomy
|
Paper:
|
Recording:
|
A General Framework for Pairwise Unbiased Learning to Rank
|
Paper:
|
Recording:
|
Reduce, Reuse, Recycle: Green Information Retrieval Research
|
Paper:
|
Recording:
|
Curriculum Learning for Dense Retrieval Distillation
|
Paper:
|
Recording:
|
Towards Validating Long-Term User Feedbacks in Interactive Recommendation Systems
|
Paper:
|
Recording:
|
Retrieval-Enhanced Machine Learning
|
Paper:
|
Recording:
|
Learning to Rank Instant Search Results with Multiple Indices: A Case Study in Search Aggregation for Entertainment
|
Paper:
|
Recording:
|
Deployable and Continuable Meta-learning-Based Recommender System with Fast User-Incremental Updates
Note: academic. not actual deployment.
|
Paper:
|
Recording:
|
Tutorials
Sequential/Session-Based Recommendations: Challenges, Approaches, Applications and Opportunities
Recording:
Self-Supervised Learning for Recommender Systems
Recording:
Recent Advances in Retrieval-Augmented Text Generation
Recording:
Beyond Opinion Mining: Summarizing Opinions of Customer Reviews
Recording:
Deep Knowledge Graph Representation Learning for Completion, Alignment and Question Answering
Recording:
Improving Efficiency and Robustness of Transformer-based Information Retrieval Systems
Note: Extremely basic info on Neural IR.
Recording: Session 1, Session 2
Conversational Information Seeking: Theory and Behaviour
Recording: Session 1, Session 2, Session 3, Session 4
Towards Reproducible Machine Learning Research in Information Retrieval
Recording: Session 1, Session 2
Gender Fairness in Information Retrieval Systems
Recording: Session 1, Session 2
Retrieval and Recommendation Systems at the Crossroads of Artificial Intelligence, Ethics, and Regulation
Recording: Session 1, Session 2
Workshops
ReNeuIR - Reaching Efficiency in Neural Information Retrieval
Recordings: Session 1 - Introduction, Session 2, Session 3 - Paper Talks, Session 4 - Round Table
DRL4IR - Reinforcement Learning for Information Retrieval
Recordings: Session 1, Session 2
SCAI - Search-Oriented Conversational Artificial Intelligence
Recordings: Session 1, Session 2, Session 3, Session 4
QUARE - Measuring the Quality of Explanations in Recommender Systems
Recordings: Session 1
IWILDS - International Workshop on Investigating Learning During Web Search
Recordings: Session 1, Session 2
ECOM - eCommerce
Recordings: Session 1, Session 2, Session 3, Session 4
INRA - News Recommendation and Analytics
Recordings: Session 1