AutoLike frames recommendation auditing as reinforcement learning and shows that automated like/skip interactions can push TikTok's feed to serve about 2x more content on a chosen topic and sentiment than a skip-only control.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AutoLike: Auditing Social Media Recommendations through User Interactions
AutoLike frames recommendation auditing as reinforcement learning and shows that automated like/skip interactions can push TikTok's feed to serve about 2x more content on a chosen topic and sentiment than a skip-only control.