CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to control attributes of Language Models (LMs) for text generation are not safe, as toxicity and bias goals are opposed to each other. |
| Approach: | They propose a method to control the attributes of Language Models (LMs) for the text generation task using Causal Average Treatment Effect (ATE) scores and counterfactual augmentation. |
| Outcome: | The proposed architecture achieves state of the art performance for toxic degeneration, which are computed using Real Toxicity Prompts. |
Similar Papers
Detoxifying Language Models Risks Marginalizing Minority Voices (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing detoxification techniques have been proposed to mitigate toxic LM generations . e.g., detoxification makes LMs more brittle to distribution shift, especially on language used by marginalized groups . |
| Approach: | They propose to use detoxification techniques to reduce toxic LM generations without affecting perplexity or generation quality on nontoxic inputs. |
| Outcome: | The proposed methods hurt equity on language used by marginalized groups, the authors show . they show that detoxification makes LMs more brittle to distribution shift, they say . |
Language Model Based Text-to-Audio Generation: Anti-Causally Aligned Collaborative Residual Transformers (2025.emnlp-main)
Copied to clipboard
| Challenge: | Autoregressive language models excel in text-to-audio generation, but lag behind diffusion models by a non-trivial margin. |
| Approach: | They propose a framework that integrates multiple isolated transformers with causal conditioning and anti-causal alignment via reinforcement learning. |
| Outcome: | The proposed framework outperforms existing LM-based and diffusion-based systems in audio synthesis. |
Contrastive Perplexity for Controlled Generation: An Application in Detoxifying Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to generate toxic content by large language models are based on pipelines . current approaches focus on preserving performance while effectively mitigating toxicity . |
| Approach: | They propose a framework for implicit knowledge editing and controlled text generation by using hard negatives. |
| Outcome: | The proposed framework significantly reduces toxic generation while maintaining strong performance on downstream tasks. |
Self-Detoxifying Language Models via Toxification Reversal (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to generate toxic content in pretrained language models are resource-intensive and require additional components. |
| Approach: | They propose a method that enables the PLM itself to achieve "self-detoxification" they identify the toxification direction from the normal generation process to the one prompted with the negative prefix and then steer the generation to the reverse direction by manipulating the information movement within the attention layers. |
| Outcome: | The proposed method can achieve comparable performance with state-of-the-art methods without any fine-tuning or extra components. |
Fine-Grained Controllable Text Generation Using Non-Residual Prompting (2022.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to control the text generation process are not expressive enough. |
| Approach: | They propose an encoder-decoder architecture that enables intermediate text prompts at arbitrary time steps. |
| Outcome: | The proposed architecture is expressive and versatile on multiple experimental settings. |
Attribute Controlled Fine-tuning for Large Language Models: A Case Study on Detoxification (2024.findings-emnlp)
Copied to clipboard
Tao Meng, Ninareh Mehrabi, Palash Goyal, Anil Ramakrishna, Aram Galstyan, Richard Zemel, Kai-Wei Chang, Rahul Gupta, Charith Peris
| Challenge: | Using a sequence-level constraint, we regularize the LLMtraining by penalizing the KL divergence between the desired output distribution and the LRM’s posterior. |
| Approach: | They propose a constraint learning schema forfine-tuning Large Language Models with attribute control by penalizing the KL divergence be-tween the desired output distribution and the LLM's posterior. |
| Outcome: | The proposed approach improves the performance of large language models while enhancing their utility and generation quality. |
COFFEE: Counterfactual Fairness for Personalized Text Generation in Explainable Recommendation (2023.emnlp-main)
Copied to clipboard
Nan Wang, Qifan Wang, Yi-Chia Wang, Maziar Sanjabi, Jingzhou Liu, Hamed Firooz, Hongning Wang, Shaoliang Nie
| Challenge: | Personalized text generation (PTG) is a key component of our digital lives but can inadvertently associate different levels of linguistic quality with users’ protected attributes. |
| Approach: | They propose a framework to achieve measure-specific counterfactual fairness in explanation generation by focusing on one of the most studied settings: generating natural language explanations for recommendations. |
| Outcome: | The proposed framework achieves measure-specific counterfactual fairness in explanation generation. |
CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) frequently generate toxic content, posing significant risks for safe deployment. |
| Approach: | They propose a framework that identifies and intervenes on the specific attention heads causally responsible for toxic generation. |
| Outcome: | The proposed framework reduces toxic generation by 5.34% while preserving linguistic fluency and speeding up head selection. |
Plug-in Language Model: Controlling Text Generation with a Simple Regression Model (2024.findings-naacl)
Copied to clipboard
| Challenge: | Large-scale pre-trained language models have demonstrated unrivaled capacity in generating text that closely resembles human-written content. |
| Approach: | They propose a plug-in language model that leverages reinforcement learning to adjust latent states to control text generation. |
| Outcome: | The proposed model outperforms existing methods that rely on gradient-based, weighted decoding, or prompt-based methods. |
Cross-Lingual Transfer of Debiasing and Detoxification in Multilingual LLMs: An Extensive Investigation (2025.findings-acl)
Copied to clipboard
| Challenge: | Prior work has shown that finetuning on specialized datasets can mitigate this behavior, and doing so in English can transfer to other languages. |
| Approach: | They propose to fine tune generative large language models to provide safe responses to harmful user input and to use direct preference optimization to mitigate toxicity. |
| Outcome: | The proposed models show that finetuning on specialized datasets reduces biases but also produces fluent and diverse text in non-English languages. |