# Wait! There's a Way Out: A Decision Mechanism for Forecasting Conversational Derailment

arXiv: https://arxiv.org/abs/2605.29243 · HTML: https://mlsanity.com/paper/2605.29243/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 28 May 2026.

## Builds on (most cited first, 12 of 12)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Conversations Gone Awry: Detecting Early Signs of Conversational Failure](https://mlsanity.com/paper/1805.05345/) (2018)
- [A Just and Comprehensive Strategy for Using NLP to Address Online Abuse](https://mlsanity.com/paper/1906.01738/) (2019)
- [Trouble on the Horizon: Forecasting the Derailment of Online Conversations as they Develop](https://mlsanity.com/paper/1909.01362/) (2019)
- [Before Name-calling: Dynamics and Triggers of Ad Hominem Fallacies in Web Argumentation](https://mlsanity.com/paper/1802.06613/) (2018)
- [Conversations Gone Alright: Quantifying and Predicting Prosocial Outcomes in Online Conversations](https://mlsanity.com/paper/2102.08368/) (2021)
- [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)
- [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)
- [Hanging in the Balance: Pivotal Moments in Crisis Counseling Conversations](https://mlsanity.com/paper/2506.03941/) (2025)
- [Conversations Gone Awry, But Then? Evaluating Conversational Forecasting Models](https://mlsanity.com/paper/2507.19470/) (2025)

## Abstract (arXiv)

Forecasting conversational derailment is the task of predicting, as the conversation unfolds, whether it will eventually derail into personal attacks. Since forecasting models operate in an online fashion, they must decide whether to "trigger" an alert after each utterance--for example, to notify participants or a moderator that the conversation is at risk of derailing. Existing approaches make this decision solely based on the estimated likelihood of derailment given the preceding utterances, implicitly assuming that the conversation's future trajectory is fixed. As a result, they ignore the possibility of future recovery and incur an unnecessarily high rate of false positives. In this work we propose a method for decoupling the decision to trigger from derailment likelihood estimation. Our approach is inspired by the first human baseline on this task, which shows that humans achieve dramatically lower false positive rates by selectively deferring their decision to trigger when they anticipate that tension is likely to subside. We operationalize this insight with a deferral mechanism that uses forward-looking simulations to assess whether a tense moment admits plausible paths to recovery. Incorporating this mechanism into a state-of-the-art forecasting model substantially reduces false positives without sacrificing forecasting accuracy. More broadly, this work highlights the value of treating decision-making as a first-class component of forecasting systems.