# Chain-of-Thought Embeddings for Stance Detection on Social Media

arXiv: https://arxiv.org/abs/2310.19750 · HTML: https://mlsanity.com/paper/2310.19750/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Causal reasoning in LLMs, one of 17 directions in LLM reasoning.
- Builds on 7 of its 17 references that are in the map.
- First posted on arXiv 30 Oct 2023.

## Placed in

- LLM reasoning / [Causal reasoning in LLMs](https://mlsanity.com/lr/causal-reasoning-in-llms/)

## Builds on (most cited first, 7 of 7)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [How would Stance Detection Techniques Evolve after the Launch of ChatGPT?](https://mlsanity.com/paper/2212.14548/) (2022)
- [Investigating Chain-of-thought with ChatGPT for Stance Detection on Social Media](https://mlsanity.com/paper/2304.03087/) (2023)
- [Infusing Knowledge from Wikipedia to Enhance Stance Detection](https://mlsanity.com/paper/2204.03839/) (2022)

## Built on by (most cited first, 13 of 13)

- [Multi-modal Stance Detection: New Datasets and Model](https://mlsanity.com/paper/2402.14298/) (2024)
- [LLM Chain Ensembles for Scalable and Accurate Data Annotation](https://mlsanity.com/paper/2410.13006/) (2024)
- [Large Language Models Meet Stance Detection: A Survey of Tasks, Methods, Applications, Challenges and Future Directions](https://mlsanity.com/paper/2505.08464/) (2025)
- [Stance Detection on Social Media with Fine-Tuned Large Language Models](https://mlsanity.com/paper/2404.12171/) (2024)
- [Stanceformer: Target-Aware Transformer for Stance Detection](https://mlsanity.com/paper/2410.07083/) (2024)
- [DIVERSE: A Dataset of YouTube Video Comment Stances with a Data Programming Model](https://mlsanity.com/paper/2403.03334/) (2024)
- [Journalism-Guided Agentic In-Context Learning for News Stance Detection](https://mlsanity.com/paper/2507.11049/) (2025)
- [EMBRACE: Shaping Inclusive Opinion Representation by Aligning Implicit Conversations with Social Norms](https://mlsanity.com/paper/2507.20264/) (2025)
- [Content Fuzzing for Escaping Information Cocoons on Digital Social Media](https://mlsanity.com/paper/2604.05461/) (2026)
- [Contextualized Prompting For Stance Detection On Social Media](https://mlsanity.com/paper/2606.06022/) (2026)
- [SICI: A Semantic-Pragmatic Complexity Index Reveals Regime Shifts in LLM Stance Detection](https://mlsanity.com/paper/2606.13189/) (2026)
- [Not All or None: Dynamic Construction of Target-aware Memory Graph for Conversational Stance Detection](https://mlsanity.com/paper/2608.29066/) (2026)
- [Visual Framing for News Stance Detection via Image Generation](https://mlsanity.com/paper/2609.00685/) (2026)

## Abstract (arXiv)

Stance detection on social media is challenging for Large Language Models (LLMs), as emerging slang and colloquial language in online conversations often contain deeply implicit stance labels. Chain-of-Thought (COT) prompting has recently been shown to improve performance on stance detection tasks -- alleviating some of these issues. However, COT prompting still struggles with implicit stance identification. This challenge arises because many samples are initially challenging to comprehend before a model becomes familiar with the slang and evolving knowledge related to different topics, all of which need to be acquired through the training data. In this study, we address this problem by introducing COT Embeddings which improve COT performance on stance detection tasks by embedding COT reasonings and integrating them into a traditional RoBERTa-based stance detection pipeline. Our analysis demonstrates that 1) text encoders can leverage COT reasonings with minor errors or hallucinations that would otherwise distort the COT output label. 2) Text encoders can overlook misleading COT reasoning when a sample's prediction heavily depends on domain-specific patterns. Our model achieves SOTA performance on multiple stance detection datasets collected from social media.