PopALM: Popularity-Aligned Language Models for Social Media Trendy Response Prediction
This addresses the need for automated prediction of mainstream public reactions on social media platforms, though it appears incremental by building on existing language models and reinforcement learning techniques.
The paper tackles the problem of predicting popular user responses to social media events by proposing Popularity-Aligned Language Models (PopALM), which uses reinforcement learning with curriculum learning to align responses with user likes, resulting in improved performance on a large-scale Weibo dataset.
Social media platforms are daily exhibiting millions of events. To preliminarily predict the mainstream public reaction to these events, we study trendy response prediction to automatically generate top-liked user replies to social media events. While previous works focus on generating responses without factoring in popularity, we propose Popularity-Aligned Language Models (PopALM) to distinguish responses liked by a larger audience through reinforcement learning. Recognizing the noisy labels from user "likes", we tailor-make curriculum learning in proximal policy optimization (PPO) to help models capture the essential samples for easy-to-hard training. In experiments, we build a large-scale Weibo dataset for trendy response prediction, and its results show that PopALM can help boost the performance of advanced language models.