An official website of the United States government
Language models have shown promise in various tasks but can be affected by undesired data during training, fine-tuning, or alignment. For example, if some unsafe conversations are wrongly annotated as safe ones, the model fine-tuned on these samples may be harmful. Therefore, the correctness of annotations, i.e., the credibility of the dataset, is important. This study focuses on the credibility of real-world datasets, including the popular benchmarks Jigsaw Civil Comments, Anthropic Harmless & Red Team, PKU BeaverTails & SafeRLHF, that can be used for training a harmless language model. Given the cost and difficulty of cleaning these datasets by humans, we introduce a systematic framework for evaluating the credibility of datasets, identifying label errors, and evaluating the influence of noisy labels in the curated language data, specifically focusing on unsafe comments and conversation classification. With the framework, we find and fix an average of 6.16% label errors in 11 datasets constructed from the above benchmarks. The data credibility and downstream learning performance can be remarkably improved by directly fixing label errors, indicating the significance of cleaning existing real-world datasets.
more »
« less
Compositional Data Augmentation for Abstractive Conversation Summarization
Recent abstractive conversation summarization systems generally rely on large-scale datasets with annotated summaries. However, collecting and annotating these conversations can be a time-consuming and labor-intensive task. To address this issue, in this work, we present a sub-structure level compositional data augmentation method, COMPO, for generating diverse and high-quality pairs of conversations and summaries. Specifically, COMPO first extracts conversation structures like topic splits and action triples as basic units. Then we organize these semantically meaningful conversation snippets compositionally to create new training instances. Additionally, we explore noise-tolerant settings in both self-training and joint-training paradigms to make the most of these augmented samples. Our experiments on benchmark datasets, SAMSum and DialogSum, show that COMPO substantially outperforms prior baseline methods by achieving a nearly 10% increase of ROUGE scores with limited data.
more »
« less
- PAR ID:
- 10467072
- Publisher / Repository:
- Association for Computational Linguistics
- Date Published:
- Page Range / eLocation ID:
- 1471 to 1488
- Subject(s) / Keyword(s):
- abstractive conversation summarization, sub-structure level compositional data augmentation, conversation structure, conversation snippets, self-training and joint-training paradigms, conversational AI
- Format(s):
- Medium: X
- Location:
- Toronto, Canada
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
We study generating abstractive summaries that are faithful and factually consistent with the given articles. A novel contrastive learning formulation is presented, which leverages both reference summaries, as positive training data, and automatically generated erroneous summaries, as negative training data, to train summarization systems that are better at distinguishing between them. We further design four types of strategies for creating negative samples, to resemble errors made commonly by two state-of-the-art models, BART and PEGASUS, found in our new human annotations of summary errors. Experiments on XSum and CNN/Daily Mail show that our contrastive learning framework is robust across datasets and models. It consistently produces more factual summaries than strong comparisons with post error correction, entailmentbased reranking, and unlikelihood training, according to QA-based factuality evaluation. Human judges echo the observation and find that our model summaries correct more errors.more » « less
-
null (Ed.)Modelling persuasion strategies as predictors of task outcome has several real-world applications and has received considerable attention from the computational linguistics community. However, previous research has failed to account for the resisting strategies employed by an individual to foil such persuasion attempts. Grounded in prior literature in cognitive and social psychology, we propose a generalised framework for identifying resisting strategies in persuasive conversations. We instantiate our framework on two distinct datasets comprising persuasion and negotiation conversations. We also leverage a hierarchical sequence-labelling neural architecture to infer the aforementioned resisting strategies automatically. Our experiments reveal the asymmetry of power roles in non-collaborative goal-directed conversations and the benefits accrued from incorporating resisting strategies on the final conversation outcome. We also investigate the role of different resisting strategies on the conversation outcome and glean insights that corroborate with past findings. We also make the code and the dataset of this work publicly available at this https URL.more » « less
-
Ensuring the online safety of youth has motivated research towards the development of machine learning (ML) methods capable of accurately detecting social media risks after-the-fact. However, for these detection models to be effective, they must proactively identify high-risk scenarios (e.g., sexual solicitations, cyberbullying) to mitigate harm. This `real-time' responsiveness is a recognized challenge within the risk detection literature. Therefore, this paper presents a novel two-level framework that first uses reinforcement learning to identify conversation stop points to prioritize messages for evaluation. Then, we optimize state-of-the-art deep learning models to accurately categorize risk priority (low, high). We apply this framework to a time-based simulation using a rich dataset of 23K private conversations with over 7 million messages donated by 194 youth (ages 13-21). We conducted an experiment comparing our new approach to a traditional conversation-level baseline. We found that the timeliness of conversations significantly improved from over 2 hours to approximately 16 minutes with only a slight reduction in accuracy (0.88 to 0.84). This study advances real-time detection approaches for social media data and provides a benchmark for future training reinforcement learning that prioritizes the timeliness of classifying high-risk conversations.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
