# Recourse for reclamation: Chatting with generative language models

arXiv: https://arxiv.org/abs/2403.14467 · HTML: https://mlsanity.com/paper/2403.14467/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 90 references that are in the map.
- First posted on arXiv 21 Mar 2024.

## Builds on (most cited first, 27 of 27)

- [Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR](https://mlsanity.com/paper/1711.00399/) (2017)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Artificial Intelligence, Values and Alignment](https://mlsanity.com/paper/2001.09768/) (2020)
- [GeDi: Generative Discriminator Guided Sequence Generation](https://mlsanity.com/paper/2009.06367/) (2020)
- [Persistent Anti-Muslim Bias in Large Language Models](https://mlsanity.com/paper/2101.05783/) (2021)
- [Actionable Recourse in Linear Classification](https://mlsanity.com/paper/1809.06514/) (2018)
- [Strategic Classification](https://mlsanity.com/paper/1506.06980/) (2015)
- [Algorithmic Recourse: from Counterfactual Explanations to Interventions](https://mlsanity.com/paper/2002.06278/) (2020)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [Challenges in Detoxifying Language Models](https://mlsanity.com/paper/2109.07445/) (2021)
- [Power to the People? Opportunities and Challenges for Participatory AI](https://mlsanity.com/paper/2209.07572/) (2022)
- [Strategic Classification from Revealed Preferences](https://mlsanity.com/paper/1710.07887/) (2017)
- [Detoxifying Language Models Risks Marginalizing Minority Voices](https://mlsanity.com/paper/2104.06390/) (2021)
- [The Social Cost of Strategic Classification](https://mlsanity.com/paper/1808.08460/) (2018)
- [Towards Robust and Reliable Algorithmic Recourse](https://mlsanity.com/paper/2102.13620/) (2021)
- [Contextualizing Hate Speech Classifiers with Post-hoc Explanation](https://mlsanity.com/paper/2005.02439/) (2020)
- [The Disparate Effects of Strategic Manipulation](https://mlsanity.com/paper/1808.08646/) (2018)
- [Contrastive Explanation: A Structural-Model Approach](https://mlsanity.com/paper/1811.03163/) (2018)
- [AI's Regimes of Representation: A Community-centered Study of Text-to-Image Models in South Asia](https://mlsanity.com/paper/2305.11844/) (2023)
- [Fairness for Unobserved Characteristics: Insights from Technological Impacts on Queer Communities](https://mlsanity.com/paper/2102.04257/) (2021)
- [A Framework of Severity for Harmful Content Online](https://mlsanity.com/paper/2108.04401/) (2021)
- [Indian-BhED: A Dataset for Measuring India-Centric Biases in Large Language Models](https://mlsanity.com/paper/2309.08573/) (2023)
- [Warmth and competence in human-agent cooperation](https://mlsanity.com/paper/2201.13448/) (2022)
- [Human participants in AI research: Ethics and transparency in practice](https://mlsanity.com/paper/2311.01254/) (2023)
- [Algorithmic Censoring in Dynamic Learning Systems](https://mlsanity.com/paper/2305.09035/) (2023)
- [A Justice-Based Framework for the Analysis of Algorithmic Fairness-Utility Trade-Offs](https://mlsanity.com/paper/2206.02891/) (2022)
- [Actionable Recourse via GANs for Mobile Health](https://mlsanity.com/paper/2211.06525/) (2022)

## Abstract (arXiv)

Researchers and developers increasingly rely on toxicity scoring to moderate generative language model outputs, in settings such as customer service, information retrieval, and content generation. However, toxicity scoring may render pertinent information inaccessible, rigidify or "value-lock" cultural norms, and prevent language reclamation processes, particularly for marginalized people. In this work, we extend the concept of algorithmic recourse to generative language models: we provide users a novel mechanism to achieve their desired prediction by dynamically setting thresholds for toxicity filtering. Users thereby exercise increased agency relative to interactions with the baseline system. A pilot study (n = 30) supports the potential of our proposed recourse mechanism, indicating improvements in usability compared to fixed-threshold toxicity-filtering of model outputs. Future work should explore the intersection of toxicity scoring, model controllability, user agency, and language reclamation processes -- particularly with regard to the bias that many communities encounter when interacting with generative language models.