Proactive_Mitigation_Detox_ICWSM

2 minutes pitching ideas at the Detox workshop ICWSM'25.

Transcript

1. User Driven Proactive Toxicity Mitigation On Web Sarah Masud Postdoc @ CopeNLU, DIKU Pioneer Center for AI Pitching @ Detox,ICWSM’25
2. Motivation For Proactive Mitigation • Howard’s The Art of Hosting Good Conversations Online • Social media punishment does not need to be a Kafkaesque nightmare • Reconsidering Tweets: Intervening during Tweet Creation Decreases Offensive Content, Katsaros et al., ICWSM ‘22 “The intervention has to be ground-up, not top-down.” - Howard Rheingold, Author
3. Our Proposed Setup For Proactive Mitigation Extremely Hateful Post (ORIGINAL) Less Hateful Post (SUGGESTIVE) HATE INTENSITY REDUCTION Extremely Hateful Post (ORIGINAL) User’s choice before posting For a given user initiated post t, with hate intensity 𝜙 𝑡 our objective is to obtain its normalized/sensitised/less intense form 𝑡` before it is posted, such that the intensity of hatred 𝜙 is reduced while the meaning still conveys. 𝜙 𝑡` < 𝜙 𝑡 OR • Proactively Reducing the Hate Intensity of Online Posts via Hate Speech Normalization, KDD 2022 User’s draft post
4. Read About Our POC Paper Link _themessier on Bluesky