Search

Open-domain dialogue systems have started to engage in continuous conversations with humans. Those dialogue systems are required to be adjusted to the human interlocutor and evaluated in terms of their perspective. However, it is questionable whether the current automatic evaluation methods can approximate the interlocutor's judgments. In this study, we analyzed and examined what features are needed in an automatic response evaluator from the interlocutor's perspective. The first experiment on the Hazumi dataset revealed that interlocutor awareness plays a critical role in making automatic response evaluation correlate with the interlocutor's judgments. The second experiment using massive conversations on X (formerly Twitter) confirmed that dialogue continuity prediction can train an interlocutor-aware response evaluator without human feedback while revealing the difficulty in evaluating generated responses compared to human responses., Comment: 9 pages, 3 figures, 5 tables, Accepted by IJCNLP-AACL 2023 SRW

The meanings of words and phrases depend not only on where they are used (contexts) but also on who use them (writers). Pretrained language models (PLMs) are powerful tools for capturing context, but they are typically pretrained and fine-tuned for universal use across different writers. This study aims to improve the accuracy of text understanding tasks by personalizing the fine-tuning of PLMs for specific writers. We focus on a general setting where only the plain text from target writers are available for personalization. To avoid the cost of fine-tuning and storing multiple copies of PLMs for different users, we exhaustively explore using writer-specific prompts to personalize a unified PLM. Since the design and evaluation of these prompts is an underdeveloped area, we introduce and compare different types of prompts that are possible in our setting. To maximize the potential of prompt-based personalized fine-tuning, we propose a personalized intermediate learning based on masked language modeling to extract task-independent traits of writers' text. Our experiments, using multiple tasks, datasets, and PLMs, reveal the nature of different prompts and the effectiveness of our intermediate learning approach., Comment: 11 pages

We make decisions by reacting to changes in the real world, in particular, the emergence and disappearance of impermanent entities such as events, restaurants, and services. Because we want to avoid missing out on opportunities or making fruitless actions after they have disappeared, it is important to know when entities disappear as early as possible. We thus tackle the task of detecting disappearing entities from microblogs, whose posts mention various entities, in a timely manner. The major challenge is detecting uncertain contexts of disappearing entities from noisy microblog posts. To collect these disappearing contexts, we design time-sensitive distant supervision, which utilizes entities from the knowledge base and time-series posts, for this task to build large-scale Twitter datasets\footnote{We will release the datasets (tweet IDs) used in the experiments to promote reproducibility.} for English and Japanese. To ensure robust detection in noisy environments, we refine pretrained word embeddings of the detection model on microblog streams of the target day. Experimental results on the Twitter datasets confirmed the effectiveness of the collected labeled data and refined word embeddings; more than 70\% of the detected disappearing entities in Wikipedia are discovered earlier than the update on Wikipedia, and the average lead-time is over one month.

Vaccines are promising tools to control the spread of COVID-19. An effective vaccination campaign requires government policies and community engagement, sharing experiences for social support, and voicing concerns to vaccine safety and efficiency. The increasing use of online social platforms allows us to trace large-scale communication and infer public opinion in real-time. We collected more than 100 million vaccine-related tweets posted by 8 million users and used the Latent Dirichlet Allocation model to perform automated topic modeling of tweet texts during the vaccination campaign in Japan. We identified 15 topics grouped into 4 themes on Personal issue, Breaking news, Politics, and Conspiracy and humour. The evolution of the popularity of themes revealed a shift in public opinion, initially sharing the attention over personal issues (individual aspect), collecting information from the news (knowledge acquisition), and government criticisms, towards personal experiences once confidence in the vaccination campaign was established. An interrupted time series regression analysis showed that the Tokyo Olympic Games affected public opinion more than other critical events but not the course of the vaccination. Public opinion on politics was significantly affected by various events, positively shifting the attention in the early stages of the vaccination campaign and negatively later. Tweets about personal issues were mostly retweeted when the vaccination reached the younger population. The associations between the vaccination campaign stages and tweet themes suggest that the public engagement in the social platform contributed to speedup vaccine uptake by reducing anxiety via social learning and support.

In this paper, we show a mean convergence theorem for a mapping with an attractive point in a Hilbert space by using a quasinonexpansive extension of the mapping and a mean convergence theorem for a quasinonexpansive mapping.