Tag
The article discusses Meta's Muse AI agent and its cute mascot Jolly, exploring how cuteness might influence user behavior and privacy, while highlighting its integration with smart glasses and the new Charm wearable.
Snapchat has built a subscription business worth over $1 billion primarily by allowing users to designate a 'Best Friend' and see who rewatch their stories, highlighting high willingness to pay for features that reveal others' feelings about you.
The paper diagnoses a systematic gap between AI preferences and real user engagement in content generation, finding that LLMs overemphasize logical structure while real engagement favors affective and expressive elements, and proposes OMRA to reduce this gap by 54.4%.
The article offers a tip to improve ChatGPT's response quality by adding a specific sentence, aimed at helping users have more effective AI conversations.
Astra, an AI assistant, detects when a user is distracted and uses an auditory cue on a Mac to bring their attention back.
Snapchat is launching Plans, a new feature that helps users organize events and meetups by setting details, inviting friends, and tracking RSVPs, aiming to boost user retention and engagement.
Tesla is requesting users to share videos of accidents prevented using their Full Self-Driving Supervised feature.
Christian Heilmann critiques the transformation of social media into a tool for distraction and rage, arguing that platforms have abandoned their social and creative roots due to engagement-driven metrics and lock-in.
This paper proposes GOAL, a constraint-aware generative framework for incentivized advertising that formulates incentive allocation as conditional sequence generation, and introduces SCPO to learn a single generative policy that generalizes across ROI constraints. Experiments show improved long-term revenue and user retention with reduced ROI violations.
Removing the top 30 highest-paid revenue share accounts from X's For You timeline increased time spent and daily active users in a 3% experiment.
A study by Wu and Holmes with 399 U.S. participants found that songs labeled as AI-generated received 23% lower emotional resonance ratings and 19% less listening time, despite most listeners being unable to accurately distinguish AI from human-made music.