# Trouble on the Horizon: Forecasting the Derailment of Online Conversations as they Develop

arXiv: https://arxiv.org/abs/1909.01362 · HTML: https://mlsanity.com/paper/1909.01362/

## Key facts

- Cited by 41 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Sep 2019.

## Built on by (most cited first, 30 of 41)

- [Conversations Gone Alright: Quantifying and Predicting Prosocial Outcomes in Online Conversations](https://mlsanity.com/paper/2102.08368/) (2021)
- [I Beg to Differ: A study of constructive disagreement in online conversations](https://mlsanity.com/paper/2101.10917/) (2021)
- [Detecting Community Sensitive Norm Violations in Online Conversations](https://mlsanity.com/paper/2110.04419/) (2021)
- [Accounting for Sycophancy in Language Model Uncertainty Estimation](https://mlsanity.com/paper/2410.14746/) (2024)
- [Dynamic Forecasting of Conversation Derailment](https://mlsanity.com/paper/2110.05111/) (2021)
- [Proactive Moderation of Online Discussions: Existing Practices and the Potential for Algorithmic Support](https://mlsanity.com/paper/2211.16525/) (2022)
- [BASIL: Bayesian Assessment of Sycophancy in LLMs](https://mlsanity.com/paper/2508.16846/) (2025)
- [Countering Online Hate Speech: An NLP Perspective](https://mlsanity.com/paper/2109.02941/) (2021)
- [Towards Online Multi-Modal Social Interaction Understanding](https://mlsanity.com/paper/2503.19851/) (2025)
- [Thread With Caution: Proactively Helping Users Assess and Deescalate Tension in Their Online Discussions](https://mlsanity.com/paper/2212.01401/) (2022)
- [Evaluation and Facilitation of Online Discussions in the LLM Era: A Survey](https://mlsanity.com/paper/2503.01513/) (2025)
- [Revisiting Contextual Toxicity Detection in Conversations](https://mlsanity.com/paper/2111.12447/) (2021)
- [How to disagree well: Investigating the dispute tactics used on Wikipedia](https://mlsanity.com/paper/2212.08353/) (2022)
- [Deal, or no deal (or who knows)? Forecasting Uncertainty in Conversations using Large Language Models](https://mlsanity.com/paper/2402.03284/) (2024)
- [How Did We Get Here? Summarizing Conversation Dynamics](https://mlsanity.com/paper/2404.19007/) (2024)
- [TempoFormer: A Transformer for Temporally-aware Representations in Change Detection](https://mlsanity.com/paper/2408.15689/) (2024)
- [Echoes of Discord: Forecasting Hater Reactions to Counterspeech](https://mlsanity.com/paper/2501.16235/) (2025)
- [Towards a Perspectivist Turn in Argument Quality Assessment](https://mlsanity.com/paper/2502.14501/) (2025)
- [Hanging in the Balance: Pivotal Moments in Crisis Counseling Conversations](https://mlsanity.com/paper/2506.03941/) (2025)
- [Conversation Modeling to Predict Derailment](https://mlsanity.com/paper/2303.11184/) (2023)
- [Social Orientation: A New Feature for Dialogue Analysis](https://mlsanity.com/paper/2403.04770/) (2024)
- [Designing Synthetic Discussion Generation Systems: A Case Study for Online Facilitation](https://mlsanity.com/paper/2503.16505/) (2025)
- [Conversations Gone Awry, But Then? Evaluating Conversational Forecasting Models](https://mlsanity.com/paper/2507.19470/) (2025)
- [EMSDialog: Synthetic Multi-person Emergency Medical Service Dialogue Generation from Electronic Patient Care Reports via Multi-LLM Agents](https://mlsanity.com/paper/2604.07549/) (2026)
- [Conversation Derailment Forecasting with Graph Convolutional Networks](https://mlsanity.com/paper/2306.12982/) (2023)
- [Forecasting Conversation Derailments Through Generation](https://mlsanity.com/paper/2504.08905/) (2025)
- [A Similarity Measure for Comparing Conversational Dynamics](https://mlsanity.com/paper/2507.18956/) (2025)
- [To Facilitate or not to Facilitate: Human and LLM Facilitator Tendencies in Online Discussions](https://mlsanity.com/paper/2607.28643/) (2026)
- [Hashing it Out: Predicting Unhealthy Conversations on Twitter](https://mlsanity.com/paper/2311.10596/) (2023)
- [Knowledge-Aware Conversation Derailment Forecasting Using Graph Convolutional Networks](https://mlsanity.com/paper/2408.13440/) (2024)

## Abstract (arXiv)

Online discussions often derail into toxic exchanges between participants. Recent efforts mostly focused on detecting antisocial behavior after the fact, by analyzing single comments in isolation. To provide more timely notice to human moderators, a system needs to preemptively detect that a conversation is heading towards derailment before it actually turns toxic. This means modeling derailment as an emerging property of a conversation rather than as an isolated utterance-level event. Forecasting emerging conversational properties, however, poses several inherent modeling challenges. First, since conversations are dynamic, a forecasting model needs to capture the flow of the discussion, rather than properties of individual comments. Second, real conversations have an unknown horizon: they can end or derail at any time; thus a practical forecasting model needs to assess the risk in an online fashion, as the conversation develops. In this work we introduce a conversational forecasting model that learns an unsupervised representation of conversational dynamics and exploits it to predict future derailment as the conversation develops. By applying this model to two new diverse datasets of online conversations with labels for antisocial events, we show that it outperforms state-of-the-art systems at forecasting derailment.