# Two-Stage Voting for Robust and Efficient Suicide Risk Detection on Social Media

arXiv: https://arxiv.org/abs/2510.08365 · HTML: https://mlsanity.com/paper/2510.08365/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 33 of its 70 references that are in the map.
- First posted on arXiv 9 Oct 2025.

## Builds on (most cited first, 30 of 33)

- [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)
- [Multi-Agent Collaboration Mechanisms: A Survey of LLMs](https://mlsanity.com/paper/2501.06322/) (2025)
- [DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference](https://mlsanity.com/paper/2004.12993/) (2020)
- [Transformers as Soft Reasoners over Language](https://mlsanity.com/paper/2002.05867/) (2020)
- [MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making](https://mlsanity.com/paper/2404.15155/) (2024)
- [Harnessing Multiple Large Language Models: A Survey on LLM Ensemble](https://mlsanity.com/paper/2502.18036/) (2025)
- [A Unified Approach to Routing and Cascading for LLMs](https://mlsanity.com/paper/2410.10347/) (2024)
- [Towards Interpretable Mental Health Analysis with Large Language Models](https://mlsanity.com/paper/2304.03347/) (2023)
- [Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions](https://mlsanity.com/paper/2410.02584/) (2024)
- [LLMs for Explainable AI: A Comprehensive Survey](https://mlsanity.com/paper/2504.00125/) (2025)
- [Metaphor Understanding Challenge Dataset for LLMs](https://mlsanity.com/paper/2403.11810/) (2024)
- [Risks from Language Models for Automated Mental Healthcare: Ethics and Structure for Implementation](https://mlsanity.com/paper/2406.11852/) (2024)
- [SarcasmBench: Towards Evaluating Large Language Models on Sarcasm Understanding](https://mlsanity.com/paper/2408.11319/) (2024)
- [RomeBERT: Robust Training of Multi-Exit BERT](https://mlsanity.com/paper/2101.09755/) (2021)
- [Early Exiting with Ensemble Internal Classifiers](https://mlsanity.com/paper/2105.13792/) (2021)
- [Ranked Voting based Self-Consistency of Large Language Models](https://mlsanity.com/paper/2505.10772/) (2025)
- [Just Ask One More Time! Self-Agreement Improves Reasoning of Language Models in (Almost) All Scenarios](https://mlsanity.com/paper/2311.08154/) (2023)
- [Agreement-Based Cascading for Efficient Inference](https://mlsanity.com/paper/2407.02348/) (2024)
- [Can Large Language Models Identify Implicit Suicidal Ideation? An Empirical Evaluation](https://mlsanity.com/paper/2502.17899/) (2025)
- [A Survey of Early Exit Deep Neural Networks in NLP](https://mlsanity.com/paper/2501.07670/) (2025)
- [PsyGUARD: An Automated System for Suicide Detection and Risk Assessment in Psychological Counseling](https://mlsanity.com/paper/2409.20243/) (2024)
- [Leveraging Large Language Models for Suicide Detection on Social Media with Limited Labels](https://mlsanity.com/paper/2410.04501/) (2024)
- [Large Language Models for Subjective Language Understanding: A Survey](https://mlsanity.com/paper/2508.07959/) (2025)
- [Hierarchical Neural Network Approaches for Long Document Classification](https://mlsanity.com/paper/2201.06774/) (2022)
- [A Comparative Analysis of Transformer and LSTM Models for Detecting Suicidal Ideation on Reddit](https://mlsanity.com/paper/2411.15404/) (2024)
- [Just a Scratch: Enhancing LLM Capabilities for Self-harm Detection through Intent Differentiation and Emoji Interpretation](https://mlsanity.com/paper/2506.05073/) (2025)
- [Suicidal Ideation Detection on Social Media: A Review of Machine Learning Methods](https://mlsanity.com/paper/2201.10515/) (2022)
- [On the Impact of Language Nuances on Sentiment Analysis with Large Language Models: Paraphrasing, Sarcasm, and Emojis](https://mlsanity.com/paper/2504.05603/) (2025)
- [Evaluating Transformer Models for Suicide Risk Detection on Social Media](https://mlsanity.com/paper/2410.08375/) (2024)

## Abstract (arXiv)

Suicide rates have risen worldwide in recent years, underscoring the urgent need for proactive prevention strategies. Social media provides valuable signals, as many at-risk individuals - who often avoid formal help due to stigma - choose instead to share their distress online. Yet detecting implicit suicidal ideation, conveyed indirectly through metaphor, sarcasm, or subtle emotional cues, remains highly challenging. Lightweight models like BERT handle explicit signals but fail on subtle implicit ones, while large language models (LLMs) capture nuance at prohibitive computational cost. To address this gap, we propose a two-stage voting architecture that balances efficiency and robustness. In Stage 1, a lightweight BERT classifier rapidly resolves high-confidence explicit cases. In Stage 2, ambiguous inputs are escalated to either (i) a multi-perspective LLM voting framework to maximize recall on implicit ideation, or (ii) a feature-based ML ensemble guided by psychologically grounded indicators extracted via prompt-engineered LLMs for efficiency and interpretability. To the best of our knowledge, this is among the first works to operationalize LLM-extracted psychological features as structured vectors for suicide risk detection. On two complementary datasets - explicit-dominant Reddit and implicit-only DeepSuiMind - our framework outperforms single-model baselines, achieving 98.0% F1 on explicit cases, 99.7% on implicit ones, and reducing the cross-domain gap below 2%, while significantly lowering LLM cost.