Lune

EMNLP2021Top-tier venue

Lawyers are Dishonest? Quantifying Representational Harms in Commonsense Knowledge Resources

Ninareh Mehrabi, Pei Zhou, Fred Morstatter, Jay Pujara, Xiang Ren, Aram Galstyan

Abstract

Warning: this paper contains content that may be offensive or upsetting. Commonsense knowledge bases (CSKB) are increasingly used for various natural language processing tasks. Since CSKBs are mostly human-generated and may reflect societal biases, it is important to ensure that such biases are not conflated with the notion of commonsense. Here we focus on two widely used CSKBs, ConceptNet and GenericsKB, and establish the presence of bias in the form of two types of representational harms, overgeneralization of polarized perceptions and representation disparity across different demographic groups in both CSKBs. Next, we find similar representational harms for downstream models that use ConceptNet. Finally, we propose a filtering-based approach for mitigating such harms, and observe that our filtered-based approach can reduce the issues in both resources and models but leads to a performance drop, leaving room for future work to build fairer and stronger commonsense models.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

11 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Deep Bidirectional Language-Knowledge Graph Pretraining cites Towards Controllable Biases in Language Generation
  • Language (Technology) is Power: A Critical Survey of "Bias" in NLP cites Towards Controllable Biases in Language Generation
  • GreaseLM: Graph REASoning Enhanced Language Models cites Towards Controllable Biases in Language Generation
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers11

  1. Deep Bidirectional Language-Knowledge Graph PretrainingMichihiro Yasunaga, Antoine Bosselut, Hongyu Ren, Xikun Zhang et al.NeurIPS 2022 · 294 citations
  2. GreaseLM: Graph REASoning Enhanced Language ModelsXikun Zhang, Antoine Bosselut, Michihiro Yasunaga, Hongyu Ren et al.ICLR 2022 · 285 citations
  3. Perturbation Augmentation for Fairer NLPRebecca Qian, Candace Ross, Jude Fernandes, Eric Michael Smith et al.EMNLP 2022 · 54 citations
  4. RICA: Evaluating Robust Inference Capabilities Based on Commonsense AxiomsPei Zhou, Rahul Khanna, Seyeon Lee, Bill Yuchen Lin et al.EMNLP 2021 · 28 citations
  5. MetaMind: Modeling Human Social Thoughts with Metacognitive Multi-Agent SystemsXuanming Zhang, Yuxuan Chen, Samuel (Min-Hsuan) Yeh, Sharon LiNeurIPS 2025 · 14 citations

Builds on4

  1. Towards Controllable Biases in Language GenerationEmily Sheng, Kai-Wei Chang, Prem Natarajan, Nanyun PengEMNLP Findings 2020 · 89 citations
  2. Language (Technology) is Power: A Critical Survey of "Bias" in NLPSu Lin Blodgett, Solon Barocas, Hal Daumé III, Hanna M. WallachACL 2020 · 68 citations
  3. R^3: Reverse, Retrieve, and Rank for Sarcasm Generation with Commonsense KnowledgeTuhin Chakrabarty, Debanjan Ghosh, Smaranda Muresan, Nanyun PengACL 2020 · 58 citations
  4. Unsupervised Commonsense Question Answering with Self-TalkVered Shwartz, Peter West, Ronan Le Bras, Chandra Bhagavatula et al.EMNLP 2020 · 25 citations

Related papers

  1. RedditBias: A Real-World Resource for Bias Evaluation and Debiasing of Conversational Language ModelsSoumya Barikeri, Anne Lauscher, Ivan Vulic, Goran GlavasACL 2021
  2. Towards Understanding and Mitigating Social Biases in Language ModelsPaul Pu Liang, Chiyu Wu, Louis-Philippe Morency, Ruslan SalakhutdinovICML 2021 · 495 citations
  3. Benchmarking Commonsense Knowledge Base Population with an Effective Evaluation DatasetTianqing Fang, Weiqi Wang, Sehyun Choi, Shibo Hao et al.EMNLP 2021 · 17 citations
  4. Analyzing Stereotypes in Generative Text Inference TasksAnna Sotnikova, Yang Trista Cao, Hal Daumé III, Rachel RudingerACL Findings 2021 · 11 citations
  5. Causal-Debias: Unifying Debiasing in Pretrained Language Models and Fine-tuning via Causal Invariant LearningFan Zhou, Yuzhou Mao, Liu Yu, Yi Yang et al.ACL 2023 · 21 citations
Dusk over the sea between two cliffs drawn in fine vertical lines