Search

Information-seeking dialogue systems, including knowledge identification and response generation, aim to respond to users with fluent, coherent, and informative responses based on users’ needs, which. To tackle this challenge, we utilize data augmentation methods and several training techniques with the pre-trained language models to learn a general pattern of the task and thus achieve promising performance. In DialDoc21 competition, our system achieved 74.95 F1 score and 60.74 Exact Match score in subtask 1, and 37.72 SacreBLEU score in subtask 2. Empirical analysis is provided to explain the effectiveness of our approaches.

To diversify and enrich generated dialogue responses, knowledge-grounded dialogue has been investigated in recent years. The existing methods tackle the knowledge grounding challenge by retrieving the relevant sentences over a large corpus and augmenting the dialogues with explicit extra information. Despite their success, however, the existing works have drawbacks in inference efficiency. This paper proposes KnowExpert, a framework to bypass the explicit retrieval process and inject knowledge into the pre-trained language models with lightweight adapters and adapt to the knowledge-grounded dialogue task. To the best of our knowledge, this is the first attempt to tackle this challenge without retrieval in this task under an open-domain chit-chat scenario. The experimental results show that Knowexpert performs comparably with some retrieval-based baselines while being time-efficient in inference, demonstrating the effectiveness of our proposed method., Comment: The first two authors contribute equally; Accepted in ACL 2022 DialDoc Workshop (Best Student Paper Award)

Amid the pandemic COVID-19, the world is facing unprecedented infodemic with the proliferation of both fake and real information. Considering the problematic consequences that the COVID-19 fake-news have brought, the scientific community has put effort to tackle it. To contribute to this fight against the infodemic, we aim to achieve a robust model for the COVID-19 fake-news detection task proposed at CONSTRAINT 2021 (FakeNews-19) by taking two separate approaches: 1) fine-tuning transformers based language models with robust loss functions and 2) removing harmful training instances through influence calculation. We further evaluate the robustness of our models by evaluating on different COVID-19 misinformation test set (Tweets-19) to understand model generalization ability. With the first approach, we achieve 98.13% for weighted F1 score (W-F1) for the shared task, whereas 38.18% W-F1 on the Tweets-19 highest. On the contrary, by performing influence data cleansing, our model with 99% cleansing percentage can achieve 54.33% W-F1 score on Tweets-19 with a trade-off. By evaluating our models on two COVID-19 fake-news test sets, we suggest the importance of model generalization ability in this task to step forward to tackle the COVID-19 fake-news problem in online social media platforms.

There has been considerable progress made towards conversational models that generate coherent and fluent responses; however, this often involves training large language models on large dialogue datasets, such as Reddit. These large conversational models provide little control over the generated responses, and this control is further limited in the absence of annotated conversational datasets for attribute specific generation that can be used for fine-tuning the model. In this paper, we first propose and evaluate plug-and-play methods for controllable response generation, which does not require dialogue specific datasets and does not rely on fine-tuning a large model. While effective, the decoding procedure induces considerable computational overhead, rendering the conversational model unsuitable for interactive usage. To overcome this, we introduce an approach that does not require further computation at decoding time, while also does not require any fine-tuning of a large language model. We demonstrate, through extensive automatic and human evaluation, a high degree of control over the generated conversational responses with regard to multiple desired attributes, while being fluent., Comment: Accepted in EMNLP findings, and code available at https://github.com/andreamad8/PPCM

This article was migrated. The article was not marked as recommended. The COVID global pandemia affected all the medical educational system. In Brazil, less than one month after the first case detected, the government confirmed the existence of sustained community transmission of the virus in the country and social distance was imposed, closing all Universities. Virtual learning, through technological advancements and its multiple resources, has been considered as an alternative to mitigate such impact, but it is also possible to develop other skills and medical competencies such as empathy, finitude, ethics and other questions related to the moment in the future doctor. Finally, is important to take care of mental health of those who work in the medical profession, reinforcing the care and the need to teach academics coping strategies to face stressful situations.

The possibility of solar energy utilization is discussed based on the relationship between urban morphology and solar radiation gain on building rooftops and wall surfaces. It was found that solar radiation on the rooftops of all the buildings in Osaka is reduced to about 86.3% when shadows of surrounding buildings are considered. It is further reduced to about 78.6% when obstacles on the rooftop are added. Power generation is estimated to be 27.3 PJ/year if photovoltaic panels were introduced to all free space on the buildings in Osaka. This corresponds to 1.93 Mt-CO 2 /year, with the reduction ratio of primary energy and CO 2 emission being about 12.4% and 8.5% respectively. Solar heat generation is estimated to be 109.1 PJ/year if solar heat collectors were introduced to all free space on the buildings in Osaka. This corresponds to 5.55 Mt-CO 2 /year, with the reduction ratio of primary energy and CO 2 emissions being about 49.6% and 24.3% respectively. The relationship between street characteristics and solar radiation on roofs and walls was analyzed for a typical fine summer day. The ratio of solar radiation on wall, roof, and ground surfaces is 17.5%, 25.7%, and 56.8%, respectively, on an average. In residential and industrial areas, there are few buildings, which experience reduced solar radiation on their roofs due to lower building coverage ratio. However, in the city center and in the residential areas, several buildings experience reduced amount of solar radiation on their roofs due to the effect of shadows by surrounding buildings. In the wide north–south and east–west streets, solar radiation on the walls is not reduced, because the heights of the buildings along the main street are aligned. In the narrow north–south streets, solar radiation on the walls is reduced when the opposite building is higher than the building in question. In the narrow east–west streets, solar radiation on the south side walls is not reduced due to the high solar altitude.

The effects of seasoning treatment for fresh sugi wood meal on the mycelial growth of shiitake was investi-gated. The mycelial growth of shiitake in sugi wood meal increased to the same level as that in extract-free sugi wood meal on the 12th day and to approximately 70% of that in konara wood meal on the 28th day of the seasoning treatment. A drastic decrease in extracts of fresh sugi wood meal occurred during the early stage of treatment, and the neutral fraction of methanol extracts decreased to less than 2%. The drainage of sugi wood meal media significantly increased during the later stage of treatment. No changes in the chemical components of sugi wood meal were detected during the seasoning treatment. The mycelial growth of shiitake in the fully seasoned sugi wood meal, which was prepared for commercial mushroom cultivation, was at almost the same level as that in the extract-free sugi wood meal. All of the seasoned sugi wood meals contained a neutral fraction of less than 1%. These results suggested that seasoning treatment can promote mycelial growth of shiitake in sugi wood meal by eliminating inhibitors and improving the physical properties of sugi wood.