相关论文: Aligning NLP Models with Target Population Perspec…

Many Natural Language Processing (NLP) systems use annotated corpora for training and evaluation. However, labeled data is often costly to obtain and scaling annotation projects is difficult, which is why annotation tasks are often…

Evaluating models on datasets often fails to capture their behavior when faced with unexpected and diverse types of inputs. It would be beneficial if we could evaluate the difference between human annotation and model prediction for an…

Contextual word embeddings such as BERT have achieved state of the art performance in numerous NLP tasks. Since they are optimized to capture the statistical properties of training data, they tend to pick up on and amplify social…

Despite the promising results of current cross-lingual models for spoken language understanding systems, they still suffer from imperfect cross-lingual representation alignments between the source and target languages, which makes the…

Public opinion polling is usually done by random sampling from the entire population, treating individual opinions as independent. In the real world, individuals' opinions are often correlated, e.g., among friends in a social network. In…

Data imbalance is common in production data, where controlled production settings require data to fall within a narrow range of variation and data are collected with quality assessment in mind, rather than data analytic insights. This…

Post-training for large language models (LLMs) is constrained by the high cost of acquiring new knowledge or correcting errors and by the unintended side effects that frequently arise from retraining. To address these issues, we introduce…

We propose a novel algorithm for learning fair representations that can simultaneously mitigate two notions of disparity among different demographic subgroups in the classification setting. Two key components underpinning the design of our…

Opinion mining and demographic attribute inference have many applications in social science. In this paper, we propose models to infer daily joint probabilities of multiple latent attributes from Twitter data, such as political sentiment…

We study the potential synergy between two different NLP tasks, both confronting predicate lexical variability: identifying predicate paraphrases, and event coreference resolution. First, we used annotations from an event coreference…

Demographic information is often used to model annotator perspectives in subjective tasks such as hate speech detection, but its benefit is inconsistent: it improves performance in some settings and behaves as noise in others. This paper…

Aligning large language models (LLMs) with diverse human preferences requires pluralistic alignment, where a single model must respect the values of multiple distinct groups simultaneously. In federated reinforcement learning from human…

Real-world data is often ambiguous; for example, human annotation produces instances with multiple conflicting class labels. Partial-label learning (PLL) aims at training a classifier in this challenging setting, where each instance is…

Attention-Aware Social Choice tackles the fundamental conflict faced by some agent communities between their desire to include all members in the decision making processes and the limited time and attention that are at the disposal of the…

Aligning Large Language Models (LLMs) with human preferences is crucial in ensuring desirable and controllable model behaviors. Current methods, such as Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization…

Capitalization is an important feature in many NLP tasks such as Named Entity Recognition (NER) or Part of Speech Tagging (POS). We are trying to reproduce results of paper which shows how to mitigate a significant performance drop when…

Modern Automatic Speech Recognition (ASR) systems can achieve high performance in terms of recognition accuracy. However, a perfectly accurate transcript still can be challenging to read due to grammatical errors, disfluency, and other…

Personalizing diffusion models using limited data presents significant challenges, including overfitting, loss of prior knowledge, and degradation of text alignment. Overfitting leads to shifts in the noise prediction distribution,…

Personalizing large language models (LLMs) is important for aligning outputs with diverse user preferences, yet existing methods struggle with flexibility and generalization. We propose CoPL (Collaborative Preference Learning), a…

Instruction tuning enhances large language models (LLMs) to follow human instructions across diverse tasks, relying on high-quality datasets to guide behavior. However, these datasets, whether manually curated or synthetically generated,…