Bias Vector: Mitigating Biases in Language Models with Task Arithmetic Approach (2025.coling-main)
Copied to clipboard
| Challenge: | Using language models (LMs) has increased in use, and the use of biases and stereotypes is creating social problems. |
| Approach: | They propose a method to mitigate LM biases by continual training on biased data . they use masked language modeling to construct a Bias Vector as the difference between biased LMs and pre-trained LM weights . |
| Outcome: | The proposed method improves on the GLUE and SEAT benchmarks. |
Similar Papers
Can Machine Unlearning Reduce Social Bias in Language Models? (2024.emnlp-industry)
Copied to clipboard
Omkar Dige, Diljot Arneja, Tsz Fung Yau, Qixuan Zhang, Mohammad Bolandraftar, Xiaodan Zhu, Faiza Khattak
| Challenge: | Existing methods for mitigating bias in language models are expensive and time-consuming . comparative studies have not evaluated their respective advantages and disadvantages . |
| Approach: | They propose to use Partitioned Contrastive Gradient Unlearning and Negation via Task Vector to reduce social biases in open-source language models. |
| Outcome: | The proposed methods outperform PCGU and DPO in debiasing models . the proposed methods can be easily tuned to balance the trade-off between bias reduction and generation quality . |
Debiasing Large Language Models with Structured Knowledge (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to reduce biases in pre-training models are hampered by their performance. |
| Approach: | They propose a method that utilizes structured knowledge to mitigate bias in LLMs . their method obviates the need for training from scratch, thus offering enhanced scalability . |
| Outcome: | The proposed method outperforms state-of-the-art (SOTA) baselines in the debiasing ability. |
Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs (2026.findings-eacl)
Copied to clipboard
| Challenge: | Despite efforts to mitigate social bias in large language models, representational harms such as stereotyping continue to exist in both open and closed-source models. |
| Approach: | They propose a method to modify model activations in forward passes by applying steering vectors to a BBQ dataset and comparing their results to bias mitigation methods. |
| Outcome: | The proposed method outperforms 3 other bias mitigation methods on the BBQ dataset and shows the lowest impact on MMLU scores. |
An Empirical Survey of the Effectiveness of Debiasing Techniques for Pre-trained Language Models (2022.acl-long)
Copied to clipboard
| Challenge: | Recent work has shown pre-trained language models capture social biases from the large amounts of text they are trained on. |
| Approach: | They propose to use Counterfactual Data Augmentation, Dropout, Iterative Nullspace Projection, Self-Debias, and SentenceDebia as bias mitigation techniques to quantify their effectiveness. |
| Outcome: | The proposed techniques are Counterfactual Data Augmentation (CDA), Dropout, Iterative Nullspace Projection, Self-Debias, and SentenceDebia. |
Conceptor-Aided Debiasing of Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Pre-trained large language models reflect inherent social biases of their training corpus. |
| Approach: | They propose two methods to identify and remove the bias subspace in pre-trained large language models such as BERT and GPT by applying conceptors to a conceptor NOT operation. |
| Outcome: | The proposed method achieves state-of-the-art (SoTA) debiasing while maintaining LLMs’ performance on the GLUE benchmark. |
Likelihood-based Mitigation of Evaluation Bias in Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are widely used to evaluate natural language generation tasks as automated metrics. |
| Approach: | They propose to use LLMs to evaluate sentences with higher likelihoods and lower likelihoods to mitigate the likelihood bias. |
| Outcome: | The proposed method overrates sentences with higher likelihoods while underrating sentences with lower likelihoods. |
BiasDPO: Mitigating Bias in Language Models through Direct Preference Optimization (2024.acl-srw)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have been shown to be effective in complex language tasks, but their potential to perpetuate biases poses significant concerns. |
| Approach: | They propose a new framework employing Direct Preference Optimization to mitigate biases in LLMs. |
| Outcome: | The proposed model outperforms the baseline model on almost all bias benchmarks and achieves better performance than open-source models. |
LLMs Are Biased Towards Output Formats! Systematically Evaluating and Mitigating Output Format Bias of LLMs (2025.naacl-long)
Copied to clipboard
Do Xuan Long, Ngoc-Hai Nguyen, Tiviatis Sim, Hieu Dao, Shafiq Joty, Kenji Kawaguchi, Nancy F. Chen, Min-Yen Kan
| Challenge: | Using format-following capabilities, state-of-the-art large language models (LLMs) can be leveraged to tailor outputs to specific task formats. |
| Approach: | They propose to define a format bias evaluation metric and establish effective strategies to reduce it. |
| Outcome: | The proposed evaluation reduces the variance in ChatGPT’s performance among wrapping formats from 235.33 to 0.71 (%2) |
Uncovering Stereotypes in Large Language Models: A Task Complexity-based Approach (2024.eacl-long)
Copied to clipboard
| Challenge: | Recent Large Language Models (LLMs) have unlocked unprecedented applications of AI. |
| Approach: | They propose to use a social benchmark to evaluate the bias protection provided by Large Language Models (LLMs) with a variety of tasks with varying complexities to assess their effectiveness. |
| Outcome: | The proposed benchmark shows that both ChatGPT and GPT-4 have strong biases with respect to nationality, gender, race, and religion. |
When and Why Does Bias Mitigation Work? (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Neural models exploit shallow surface features to perform language understanding tasks, rather than learning the deeper language understanding and reasoning skills that practitioners desire. |
| Approach: | They propose to use model debiasing techniques to pressure models away from spurious features and to use them to learn useful representations instead. |
| Outcome: | The proposed methods increase models' reliance on hidden biases instead of learning robust features that help them solve a task. |