Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval
Abstract
Welcome to the 43rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2020). SIGIR is the premier scientific conference in the broad area of information retrieval. According to the original plan, SIGIR 2020 would be held at Hyatt Regency in Xi'an, China. Xi'an, known by most people as home to Terracotta Army of Emperor Qin, is one of the oldest cities in China, having held the position under 13 dynasties and being the starting point of the Silk Road. The beautiful Hyatt hotel was also carefully chosen after we had site visits for more than 15 local hotels and done frequent communications with experts and professionals across academia, government and industry. Due to the pandemic of COVID-19, the organizing committee of SIGIR 2020 decided to move this year's SIGIR fully online in April. Thus, our conference this year becomes the first virtual one in SIGIR history. This last-minute change of the conference posed many challenges and much burden in terms of preparation and sponsorships. Despite the shortened time period and challenges, General Chairs, Vice Chair and their local organizing team worked very diligently to carefully prepare an outstanding program for everyone. This year SIGIR has received a record high number of submissions in its history, as shown in the following statistics. This trend suggests a renewed interest in our field. We were happy to observe the highest number of submissions for long, short, industry track and demo papers, which sums up to 1,180 papers. The accepted papers were made by 1,221 authors from 32 countries. - 555 and 507 valid submissions for full and short papers respectively - 80 effective submissions for industry track papers plus 38 demo papers submitted - 147 papers were accepted for presentation as full papers (26.49% acceptance rate), 152 were accepted for short papers (29.98%), 22 were accepted as industry track papers (27.50%), and 19 were accepted as demo papers. In addition to above tracks, SIGIR 2020 features 6 keynote speakers, 2 invited talks for this year's industry track - Symposium on IR in Practice (SIRIP), 8 pre-conference tutorials, 8 workshops, and 14 doctoral consortium papers. We are proud of our program and acknowledge the tireless efforts of people who materialized all this together.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
MaxSim has been shown to be both effective and computationally attractive in multi-vector retrieval systems.
“This winner-take-all formulation has been shown to be both effective and computationally attractive in multi-vector retrieval systems [1, 4, 5].”
From Spike Hijacking in Late-Interaction Retrieval (2026) · cites this paper as [1]
Late-interaction retrieval models retain token or patch-level embeddings and compute relevance by aggregating fine-grained similarity scores.
“Instead of compressing a query and document into single representations, late-interaction approaches retain token or patch-level embeddings and compute relevance by aggregating fine-grained similarity scores.”
From Spike Hijacking in Late-Interaction Retrieval (2026) · cites this paper as [1]
A widely adopted aggregation rule in late-interaction retrieval is hard maximum similarity (MaxSim), where each query token contributes the maximum similarity it achieves over document patches.
“A widely adopted aggregation rule in such models is hard maximum similarity (MaxSim), where each query token contributes the maximum similarity it achieves over document patches.”
From Spike Hijacking in Late-Interaction Retrieval (2026) · cites this paper as [1]
Cites
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225 · see citing works
Power to the People: The Role of Humans in Interactive Machine Learning
Amershi, S., Cakmak, M., Knox, W. B., & Kulesza, T. (2014) · AI Magazine · doi:10.1609/aimag.v35i4.2513 · Cited by 1,084 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Cabitza, F., Campagner, A., Soares, F., García de Guadiana-Romualdo, L., Challa, F., Sulejmani, A., Seghezzi, M., & Carobene, A. (2021) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2021.106288 · Cited by 284 · see citing works
A Survey on the Memory Mechanism of Large Language Model-based Agents
Zhang, Z., Dai, Q., Bo, X., Ma, C., Li, R., Chen, X., Zhu, J., Dong, Z., & Wen, J.-R. (2025) · ACM Transactions on Information Systems · doi:10.1145/3748302 · Cited by 118 · see citing works
Self-citationFairness metrics and bias mitigation strategies for rating predictions
Ashokan, A., & Haas, C. (2021) · Information Processing & Management · doi:10.1016/j.ipm.2021.102646 · Cited by 47 · see citing works
Learning towards conversational AI: A survey
Fu, T., Gao, S., Zhao, X., Wen, J.-R., & Yan, R. (2022) · AI Open · doi:10.1016/j.aiopen.2022.02.001 · Cited by 46 · see citing works
Self-citationSpike Hijacking in Late-Interaction Retrieval
Suresh, K., Vatsa, T., King, T., Kadav, A., & Friedrich, M. (2026) · arXiv · doi:10.48550/arxiv.2604.05253
- Authors
- Huang, J., Chang, Y., Cheng, X., Kamps, J., Murdock, V., Wen, J.-R., & Liu, Y.
- Affiliations
- Université Toulouse III - Paul Sabatier (FR)
- Corresponding author
- Huang, J.
- ORCID
- Jaap Kamps
- Publisher
- ACM
- Published
- 2020-07-25
- OpenAlex
- W3201053014
- Cited by
- 722
- References
- 2
- Citations by year
- 2025: 218 · 2024: 168 · 2023: 158 · 2022: 124 · 2021: 49 · 2020: 4 · 2019: 1
- Type
- paratext
- Language
- en
- Open access
- closed
- Locations
- UvA-DARE (University of Amsterdam) (publishedVersion)
- Data Archiving and Networked Services (DANS) (submittedVersion)
- Funders
- Agence Nationale de la Recherche (ANR-11-LABX-0040)
- Topics
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Algorithms and Data Compression
- Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
- Keywords
- information retrieval
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review