Social-Group-Agnostic Bias Mitigation via the Stereotype Content Model

Omrani, Ali; S., Ziabari, Alireza; Yu, Charles; Golazizian, Preni; Kennedy, Brendan J.; Atari, Mohammad; Ji, Heng; Dehghani, Morteza

doi:10.18653/v1/2023.acl-long.227

Cited by 5 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Though modern approaches have made great advances in democratizing LLM training, most builders don't have a need to pretrain their own LLMs, opting to, at most, fine-tune them. Rather than hope that an LLM is unbiased after pretraining, many researchers have discussed the utility in having a separate general debiasing step to account for any unintended associations stemming from pretraining Omrani et al, 2023;. Relatively less explored is the complementary requirement of augmenting LLMs with the awareness and ability to abide by sociocultural norms.…”

Section: Human-centered Nlpmentioning

confidence: 99%

Defining a New NLP Playground

Li,

Han,

et al. 2023

Findings of the Association for Computational Linguistics: EMNLP 2023

View full text Add to dashboard Cite

The recent explosion of performance of large language models (LLMs) has changed the field of Natural Language Processing (NLP) more abruptly and seismically than any other shift in the field's 80-year history. This has resulted in concerns that the field will become homogenized and resource-intensive. The new status quo has put many academic researchers, especially PhD students, at a disadvantage. This paper aims to define a new NLP playground by proposing 20+ PhD-dissertation-worthy research directions, covering theoretical analysis, new and challenging problems, learning paradigms, and interdisciplinary applications.

show abstract

Section: Human-centered Nlpmentioning

confidence: 99%

Defining a New NLP Playground

Li,

Han,

et al. 2023

Findings of the Association for Computational Linguistics: EMNLP 2023

View full text Add to dashboard Cite

show abstract

“…They disentangle po-tentially correlated concepts by projecting representations into orthogonal subspaces, thus removing discriminatory correlation bias (Dev et al 2021;Kaneko and Bollegala 2021). Group-specific subspace projection requires prior group knowledge, some work (Ungless et al 2022;Omrani et al 2023) projects representations to stereotype content models (SCM) (Fiske et al 2002) that rely on theoretical understanding of social stereotypes to define bias subspaces, thus breaking the limitations of prior knowledge.…”

Section: In-processingmentioning

confidence: 99%

Rural Settlements Dataset in Changchun-Jilin Consolidated Metropolitan Area _2010

LI¹,

WANG²,

ZHANG³

et al.

GCdataPR

View full text Add to dashboard Cite

This paper concerns with the global dynamics of classical solutions to an important alarm-taxis ecosystem, which demonstrates the behaviors of prey that attract secondary predator when threatened by primary predator. And the secondary predator pursues the signal generated by the interaction of the prey and primary predator. However, it seems that the necessary gradient estimates for global existence cannot be obtained in critical case due to strong coupled structure. Thereby, we develop a new approach to estimate the gradient of prey and primary predator which takes advantage of slightly higher damping power. Then the boundedness of classical solutions in two dimension with Neumann boundary conditions can be established by energy estimates and semigroup theory. Moreover, by constructing Lyapunov functional, it is proved that the coexistence homogeneous steady states is asymptotically stability and the convergence rate is exponential under certain assumptions on the system coefficients.

show abstract

Federated Constrastive Learning and Visual Transformers for Personal Recommendation

Belhadi,

Djenouri,

de Alcantara Andrade

et al. 2024

Cogn Comput

View full text Add to dashboard Cite

This paper introduces a novel solution for personal recommendation in consumer electronic applications. It addresses, on the one hand, the data confidentiality during the training, by exploring federated learning and trusted authority mechanisms. On the other hand, it deals with data quantity, and quality by exploring both transformers and consumer clustering. The process starts by clustering the consumers into similar clusters using contrastive learning and k-means algorithm. The local model of each consumer is trained on the local data. The local models of the consumers with the clustering information are then sent to the server, where integrity verification is performed by a trusted authority. Instead of traditional federated learning solutions, two kinds of aggregation are performed. The first one is the aggregation of all models of the consumers to derive the global model. The second one is the aggregation of the models of each cluster to derive a local model of similar consumers. Both models are sent to the consumers, where each consumer decides which appropriate model might be used for personal recommendation. Robust experiments have been carried out to demonstrate the applicability of the method using MovieLens-1M, and Amazon-book. The results reveal the superiority of the proposed method compared to the baseline methods, where it reaches an average accuracy of 0.27, against the other methods that do not exceed 0.25.

show abstract

Social-Group-Agnostic Bias Mitigation via the Stereotype Content Model

Cited by 5 publications

References 0 publications

Defining a New NLP Playground

Defining a New NLP Playground

Rural Settlements Dataset in Changchun-Jilin Consolidated Metropolitan Area _2010

Federated Constrastive Learning and Visual Transformers for Personal Recommendation

Contact Info

Product

Resources

About