Are Akpans Trick or Treat: Unveiling Helpful Biases in Assistant Systems
Not on the map. Why?
Abstract (arXiv)
Information-seeking AI assistant systems aim to answer users' queries about knowledge in a timely manner. However, both the human-perceived helpfulness of information-seeking assistant systems and its fairness implication are under-explored. In this paper, we study computational measurements of helpfulness. We collect human annotations on the helpfulness of dialogue responses, develop models for automatic helpfulness evaluation, and then propose to use the helpfulness level of a dialogue system towards different user queries to gauge the fairness of a dialogue system. Experiments with state-of-the-art dialogue systems, including ChatGPT, under three information-seeking scenarios reveal that existing systems tend to be more helpful for questions regarding concepts from highly-developed countries than less-developed countries, uncovering potential fairness concerns underlying the current information-seeking assistant systems.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 30 of its 53 references that are in the map. First posted on arXiv 25 May 2022.
Builds on
30 of 53 references.
References · 30
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
- 2019ALBERT: A Lite BERT for Self-supervised Learning of Language Representationscited by 2,772
- 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,157
- 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
- 2021A General Language Assistant as a Laboratory for Alignmentcited by 1,023
- 2020Recipes for building an open-domain chatbotcited by 593
- 2019The Woman Worked as a Babysitter: On Biases in Language Generationcited by 540
- 2020A Simple Language Model for Task-Oriented Dialoguecited by 331
- 2021Towards Emotional Support Dialog Systemscited by 294
- 2016A Network-based End-to-End Trainable Task-oriented Dialogue Systemcited by 252
18 more
- 2019A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cyclecited by 192
- 2019Queens are Powerful too: Mitigating Gender Bias in Dialogue Generationcited by 152
- 2019Taskmaster-1: Toward a Realistic and Diverse Dialog Datasetcited by 139
- 2021Recent Advances in Deep Learning Based Dialogue Systems: A Systematic Surveycited by 130
- 2020UBAR: Towards Fully End-to-End Task-Oriented Dialog Systems with GPT-2cited by 111
- 2020Knowledge-Grounded Dialogue Generation with Pre-trained Language Modelscited by 107
- 2021A Comprehensive Assessment of Dialog Evaluation Metricscited by 90
- 2019Does Gender Matter? Towards Fairness in Dialogue Systemscited by 89
- 2020RobBERT: a Dutch RoBERTa-based Language Modelcited by 88
- 2018Conversational AI: The Science Behind the Alexa Prizecited by 63
- 2021Broaden the Vision: Geo-Diverse Visual Commonsense Reasoningcited by 59
- 2020Mitigating Gender Bias for Neural Dialogue Generation with Adversarial Learningcited by 44
- 2020Towards Unified Dialogue System Evaluation: A Comprehensive Analysis of Current Evaluation Protocolscited by 35
- 2021Detecting Cross-Geographic Biases in Toxicity Modeling on Social Mediacited by 31
- 2018Advancing the State of the Art in Open Domain Dialog Systems through the Alexa Prizecited by 29
- 2020Is Your Goal-Oriented Dialog Model Performing Really Well? Empirical Analysis of System-wise Evaluationcited by 19
- 2020TicketTalk: Toward human-level performance with end-to-end, transaction-based dialog systemscited by 18
- 2020Easy, Reproducible and Quality-Controlled Data Collection with Crowdaqcited by 9
Built on by
Citing papers per year: 2023: 1.
1 paper.