Robust Harmful Meme Detection under Missing Modalities via Shared Representation Learning Conference: Companion Proceedings of the ACM Web Conference 2026; April 13–17, 2026; Dubai, United Arab Emirates Companion Proceedings of the ACM Web Conference 2026 (WWW Companion ’26), April 13–17, 2026, Dubai, United Arab Emirates DOI: 10.1145/3774905.3795610 ISBN: 979-8-4007-2308-7/2026/04 CCS: Computing methodologies Neural networks
Felix Breiteneder, Mohammad Belal, Muhammad Saad Saeed, Shahed Masoudian, Usman Naseem, Kulshrestha Juhi, Markus Schedl, Shah Nawaz · 2026
Read Robust Harmful Meme Detection under Missing Modalities via Shared Representation Learning Conference: Companion Proceedings of the ACM Web Conference 2026; April 13–17, 2026; Dubai, United Arab Emirates Companion Proceedings of the ACM Web Conference 2026 (WWW Companion ’26), April 13–17, 2026, Dubai, United Arab Emirates DOI: 10.1145/3774905.3795610 ISBN: 979-8-4007-2308-7/2026/04 CCS: Computing methodologies Neural networks by Felix Breiteneder, Mohammad Belal, Muhammad Saad Saeed, Shahed Masoudian, Usman Naseem, Kulshrestha Juhi, Markus Schedl, Shah Nawaz (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-01101.txt.
Description
Felix Breiteneder and co-authors test harmful meme spotting when a sense goes missing, as when bad text reads kill the words. Most past spotters assume full text plus picture and wilt when one sense drops. Their base plan maps each sense alone into one shared space, so it still works when a sense is gone. On two test sets, it beat past spotters when text went missing, leaning more on pictures and less on words. The work marks a step toward meme spotting in real, patchy data.
Catalog record
| CREATOR | Felix Breiteneder, Mohammad Belal, Muhammad Saad Saeed, Shahed Masoudian, Usman Naseem, Kulshrestha Juhi, Markus Schedl, Shah Nawaz |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 22401 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01101