PopALM: Popularity-Aligned Language Models for Social Media Trendy Response Prediction

TL;DR

PopALM introduces a reinforcement learning approach with curriculum learning to generate social media responses that are more likely to be liked by users, improving trend prediction accuracy.

Contribution

We propose PopALM, a novel popularity-aligned language model that incorporates reinforcement learning and curriculum strategies to enhance social media response prediction.

Findings

PopALM outperforms baseline models on Weibo dataset.

Curriculum learning improves model robustness to noisy labels.

Reinforcement learning aligns responses with user preferences.

Abstract

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.