# PERCEIVE: A Benchmark for Personalized Emotion and Communication Behavior Understanding on Social Media

arXiv: https://arxiv.org/abs/2605.12525 · HTML: https://mlsanity.com/paper/2605.12525/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 38 references that are in the map.
- First posted on arXiv 10 Apr 2026.

## Builds on (most cited first, 21 of 21)

- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [GoEmotions: A Dataset of Fine-Grained Emotions](https://mlsanity.com/paper/2005.00547/) (2020)
- [LaMP: When Large Language Models Meet Personalization](https://mlsanity.com/paper/2304.11406/) (2023)
- [SemEval-2017 Task 4: Sentiment Analysis in Twitter](https://mlsanity.com/paper/1912.00741/) (2019)
- [Reasoning Implicit Sentiment with Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.11255/) (2023)
- [EmoBank: Studying the Impact of Annotation Perspective and Representation Format on Dimensional Emotion Analysis](https://mlsanity.com/paper/2205.01996/) (2022)
- [Unveiling the Truth and Facilitating Change: Towards Agent-based Large-scale Social Movement Simulation](https://mlsanity.com/paper/2402.16333/) (2024)
- [Large Language Model Hacking: Quantifying the Hidden Risks of Using LLMs for Text Annotation](https://mlsanity.com/paper/2509.08825/) (2025)
- [SimUSER: Simulating User Behavior with Large Language Models for Recommender System Evaluation](https://mlsanity.com/paper/2504.12722/) (2025)
- [LLMs + Persona-Plug = Personalized LLMs](https://mlsanity.com/paper/2409.11901/) (2024)
- [Counterfactual Debating with Preset Stances for Hallucination Elimination of LLMs](https://mlsanity.com/paper/2406.11514/) (2024)
- [Learning to Predict Persona Information forDialogue Personalization without Explicit Persona Description](https://mlsanity.com/paper/2111.15093/) (2021)
- [M-ABSA: A Multilingual Dataset for Aspect-Based Sentiment Analysis](https://mlsanity.com/paper/2502.11824/) (2025)
- [Do we still need Human Annotators? Prompting Large Language Models for Aspect Sentiment Quad Prediction](https://mlsanity.com/paper/2502.13044/) (2025)
- [From Generic Empathy to Personalized Emotional Support: A Self-Evolution Framework for User Preference Alignment](https://mlsanity.com/paper/2505.16610/) (2025)
- [Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification](https://mlsanity.com/paper/2503.15117/) (2025)
- [My Words Imply Your Opinion: Reader Agent-based Propagation Enhancement for Personalized Implicit Emotion Analysis](https://mlsanity.com/paper/2412.07367/) (2024)
- [iNews: A Multimodal Dataset for Modeling Personalized Affective Responses to News](https://mlsanity.com/paper/2503.03335/) (2025)
- [Implicit Behavioral Alignment of Language Agents in High-Stakes Crowd Simulations](https://mlsanity.com/paper/2509.16457/) (2025)
- [BeSimulator: A Large Language Model Powered Text-based Behavior Simulator](https://mlsanity.com/paper/2409.15865/) (2024)
- [Evaluating Cognitive-Behavioral Fixation via Multimodal User Viewing Patterns on Social Media](https://mlsanity.com/paper/2509.04823/) (2025)

## Abstract (arXiv)

Current emotion analysis in social media is predominantly author-centric, failing to capture the subjective nature of emotional responses across diverse readers. This paradigm overlooks the crucial link between individual perception, communication behavior, and the underlying social network. To bridge this gap, we introduce PERCEIVE, a novel bilingual (English and Chinese) large-scale benchmark that, to the best of our knowledge, is the first to integrate five critical dimensions for social perception: author-created content, genuine readers' emotional feedback (derived from their comments), communication behavior, user attributes, and the social graph. This benchmark enables a paradigm shift towards truly personalized, reader-centric analysis, where different readers' emotional responses to the same content are naturally captured through their real-world interactions. By annotating emotions from reader comments and synchronously capturing communication intent, PERCEIVE provides a unique resource to model the intrinsic coupling between emotion and behavior, grounded in social context. We establish a comprehensive evaluation protocol, testing state-of-the-art methods, including large language models (LLMs) with advanced reasoning enhancement. Our findings reveal significant shortcomings in existing approaches when handling this multifaceted, user-aware task. PERCEIVE offers a foundational resource and clear direction for future research in socially-intelligent NLP, pushing models towards a more unified understanding of emotion on social media.