An official website of the United States government
While existing work on studying bias in NLP focuses on negative or pejorative language use, Govindarajan et al. (2023) offer a revised framing of bias in terms of intergroup social context, and its effects on language behavior. In this paper, we investigate if two pragmatic features (specificity and affect) systematically vary in different intergroup contexts — thus connecting this new framing of bias to language output. Preliminary analysis finds modest correlations between specificity and affect of tweets with supervised intergroup relationship (IGR) labels. Counterfactual probing further reveals that while neural models finetuned for predicting IGR reliably use affect in classification, the model’s usage of specificity is inconclusive.
more »
« less
How people talk about each other: Modeling Generalized Intergroup Bias and Emotion
Current studies of bias in NLP rely mainly on identifying (unwanted or negative) bias towards a specific demographic group. While this has led to progress recognizing and mitigating negative bias, and having a clear notion of the targeted group is necessary, it is not always practical. In this work we extrapolate to a broader notion of bias, rooted in social science and psychology literature. We move towards predicting interpersonal group relationship (IGR) - modeling the relationship between the speaker and the target in an utterance - using fine-grained interpersonal emotions as an anchor. We build and release a dataset of English tweets by US Congress members annotated for interpersonal emotion - the first of its kind, and ‘found supervision’ for IGR labels; our analyses show that subtle emotional signals are indicative of different biases. While humans can perform better than chance at identifying IGR given an utterance, we show that neural models perform much better; furthermore, a shared encoding between IGR and interpersonal perceived emotion enabled performance gains in both tasks.
more »
« less
- PAR ID:
- 10432263
- Date Published:
- Journal Name:
- Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
- Page Range / eLocation ID:
- 2496–2506
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract The present research examined children’s anger proneness, emotion understanding, and maternal sensitivity during toddlerhood as predictors of children’s hostile attribution bias (HAB) during the later preschool years. At 2.8 years (N = 128), maternal sensitivity (e.g., child‐centered behavior) was observed during mother–child play and snack, and parents reported on children’s anger proneness. At 3.3 years, emotion understanding (i.e., ability to identify emotional expressions accurately) was measured via an interactive puppet interview. At 4.8 and 5.4 years, children's HAB was assessed via child responses to hypothetical vignettes of ambiguous peer provocations. Path models revealed that maternal sensitivity predicted fewer hostile attributions. In addition, emotion understanding and maternal sensitivity emerged as buffers against the negative effect of anger proneness on HAB. Specifically, greater anger proneness was associated with more frequent hostile attributions, but only when children had lower emotion understanding or had mothers who were less sensitive. The findings highlight the interplay between intrapersonal and interpersonal factors in early childhood that contribute to a hostile attribution bias during the preschool period.more » « less
-
Advancing speech emotion recognition (SER) de- pends highly on the source used to train the model, i.e., the emotional speech corpora. By permuting different design parameters, researchers have released versions of corpora that attempt to provide a better-quality source for training SER. In this work, we focus on studying communication modes of collection. In particular, we analyze the patterns of emotional speech collected during interpersonal conversations or monologues. While it is well known that conversation provides a better protocol for eliciting authentic emotion expressions, there is a lack of systematic analyses to determine whether conversational speech provide a “better-quality” source. Specifically, we examine this research question from three perspectives: perceptual differences, acoustic variability and SER model learning. Our analyses on the MSP- Podcast corpus show that: 1) rater’s consistency for conversation recordings is higher when evaluating categorical emotions, 2) the perceptions and acoustic patterns observed on conversations have properties that are better aligned with expected trends discussed in emotion literature, and 3) a more robust SER model can be trained from conversational data. This work brings initial evidences stating that samples of conversations may provide a better-quality source than samples from monologues for building a SER model.more » « less
-
null (Ed.)This paper proposes a metric that we call the structured saliency benchmark (SSBM) to evaluate importance maps computed for automatic speech recognizers on individual utterances. These maps indicate time-frequency points of the utterance that are most important for correct recognition of a target word. Our evaluation technique is not only suitable for standard classification tasks, but is also appropriate for structured prediction tasks like sequence-to-sequence models. Additionally, we use this approach to perform a comparison of the importance maps created by our previously introduced technique using “bubble noise” to identify important points through correlation with a baseline approach based on smoothed speech energy and forced alignment. Our results show that the bubble analysis approach is better at identifying important speech regions than this baseline on 100 sentences from the AMI corpus.more » « less
-
Prior literature has documented the importance of faculty advisors in the doctoral student socialization process, with a few studies describing negative advising relationships characterized by disengagement, disinterest, unsupportive behavior, and interpersonal conflict. We extend this research by exploring how negative advising relationships emerge and develop over time. Examining longitudinal interviews over four years with 15 doctoral students in biological sciences in the USA who experienced negative relationships with their advisors, we illuminate how negative advising relationships unfold over the course of graduate studies. We find two primary patterns in challenging relationships: some students show a gradual decline in relationship health over time, while others point to a single event altering their relationship trajectory. We also identify specific factors that shape each of these negative relationship types. By revealing the different social processes that underlie the emergence of negative advising relationships, our findings provide a valuable contribution to understanding the complex social landscape of doctoral education. The findings further the dialogue on how faculty advisors can craft successful pathways through graduate education, thereby supporting the academic and professional success of doctoral students.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
