13197 research outputs found
Sort by
VR パブリック スピーキング クンレン ニ オケル コトナル シテン ガ クンレン コウカ ニ アタエル エイキョウ オヨビ クンレン コウカ ト シンタイカ カンカク ヤ リンジョウカン トノ カンケイ
Get PDF奈良先端科学技術大学院大学修士(工学)master thesi
フェムト ビョウ レーザー ユウキ ショウゲキリョク ノ タダンカイ サヨウ ニ ヨル ビショウキュウ ウンドウ ノ カイセキ モデル
Get PDF奈良先端科学技術大学院大学修士(工学)master thesi
Gathering in Carrier Graphs: Meeting via Public Transportation System
Get PDF奈良先端科学技術大学院大学修士(理学)master thesi
InstructCMP: Length Control in Sentence Compression through Instruction-based Large Language Models
No full textExtractive summarization can produce faithful summaries but often requires additional constraints such as a desired summary length. Traditional sentence compression models do not typically consider the constraints because of their restricted model abilities, which require model modifications for coping with them. To bridge this gap, we propose Instruction-based Compression (InstructCMP), an approach to the sentence compression task that can consider the length constraint through instructions by leveraging the zero-shot task-solving abilities of Large Language Models (LLMs). For this purpose, we created new evaluation datasets by transforming traditional sentence compression datasets into an instruction format. By using the datasets, we first reveal that the current LLMs still face challenges in accurately controlling the length for a compressed text. To address this issue, we propose an approach named length priming, that incorporates additional length information into the instructions without external resources. While the length priming effectively works in a zero-shot setting, a training dataset with the instructions would further improve the ability of length control. Thus, we additionally created a training dataset in an instruction format to fine-tune the model on it. Experimental results and analysis show that applying the length priming significantly improves performances of InstructCMP in both zero-shot and fine-tuning settings without the need of any model modifications.conference pape
Towards Artwork Explanation in Large-scale Vision Language Models
No full textLarge-scale Vision-Language Models (LVLMs) output text from images and instructions, demonstrating advanced capabilities in text generation and comprehension. However, it has not been clarified to what extent LVLMs understand the knowledge necessary for explaining images, the complex relationships between various pieces of knowledge, and how they integrate these understandings into their explanations. To address this issue, we propose a new task: the artwork explanation generation task, along with its evaluation dataset and metric for quantitatively assessing the understanding and utilization of knowledge about artworks. This task is apt for image description based on the premise that LVLMs are expected to have pre-existing knowledge of artworks, which are often subjects of wide recognition and documented information.It consists of two parts: generating explanations from both images and titles of artworks, and generating explanations using only images, thus evaluating the LVLMs’ language-based and vision-based knowledge.Alongside, we release a training dataset for LVLMs to learn explanations that incorporate knowledge about artworks.Our findings indicate that LVLMs not only struggle with integrating language and visual information but also exhibit a more pronounced limitation in acquiring knowledge from images alone. The datasets ExpArt=Explain Artworks are available at https://huggingface.co/datasets/naist-nlp/ExpArtconference pape
Generating Attractive Ad Text by Facilitating the Reuse of Landing Page Expressions
No full textAd text generation is vital for automatic advertising in various fields through search engine advertising (SEA) to avoid the cost problem caused by laborious human efforts for creating ad texts. Even though ad creators create the landing page (LP) for advertising and we can expect its quality, conventional approaches with reinforcement learning (RL) mostly focus on advertising keywords rather than LP information. This work investigates and shows the effective usage of LP information as a reward in RL-based ad text generation through automatic and human evaluations. Our analysis of the actually generated ad text shows that LP information can be a crucial reward by appropriately scaling its value range to improve ad text generation performance.conference pape
Performance Evaluation of Fingerprint-Based Indoor Positioning Using RSSI in 802.11ah
No full textIn incineration plants, indoor positioning systems are needed to prevent workers from approaching hazardous areas and to facilitate rescue operations in emergencies. Although many positioning systems using signal strength indicators such as Wi-Fi Received Signal Strength Indicator (RSSI) have been proposed, the vast interior of incineration plants increases the cost of constructing a communication infrastructure using 5GHz Wi-Fi. Therefore, by adopting Low Power Wide Area (LPWA), which allows for long-distance communication, the number of Access points (APs) that need to be installed in the environment can be reduced, thus lowering the cost of constructing the communication infrastructure. IEEE 802.11ah (11ah) is a new LPWA standard that uses the Sub-GHz band. In this study, we deployed new 11ah receivers in an incineration plant where IEEE 802.11ac (11ac) is already established and evaluated their coverage and area classification performance. The experiment demonstrated that the entire incineration plant can be covered with fewer units compared to 11ac. Additionally, when a small number of 11 ah receivers were installed, the system identified individual rooms with higher accuracy than 11ac, suggesting that 11ah RSSI is a promising feature that can be obtained at low cost.conference pape
Neural End-To-End Speech Translation Leveraged by ASR Posterior Distribution
No full textEnd-to-end speech translation (ST) directly renders source language speech to the target language without intermediate automatic speech recognition (ASR) output as in a cascade approach. End-to-end ST avoids error propagation from intermediate ASR results. Although recent attempts have applied multi-task learning using an auxiliary task of ASR to improve ST performance, they use cross-entropy loss to one-hot references in the ASR task, and the trained ST models do not consider possible ASR confusion. In this study, we propose a novel multi-task learning framework for end-to-end STs leveraged by ASR-based loss against posterior distributions obtained using a pre-trained ASR model called ASR posterior-based loss (ASR-PBL). The ASR-PBL method, which enables a ST model to reflect possible ASR confusion among competing hypotheses with similar pronunciations, can be applied to one of the strong multi-task ST baseline models with Hybrid CTC/Attention ASR task loss. In our experiments on the Fisher Spanish-to-English corpus, the proposed method demonstrated better BLEU results than the baseline that used standard CE loss.journal articl
Estimation of Happiness Changes through Longitudinal Analysis of Employees' Texts
No full textMeasuring happiness as a determinant of well-being is increasingly recognized as crucial. While previous studies have utilized free-text descriptions to estimate happiness on a broad scale, limited research has focused on tracking individual fluctuations in happiness over time owing to the challenges associated with longitudinal data collection. This study addresses this issue by obtaining longitudinal data from two workplaces over two and six months respectively.Subsequently, the data is used to construct a happiness estimation model and assess individual happiness levels.Evaluation of the model performance using correlation coefficients shows variability in the correlation values among individuals.Notably, the model performs satisfactorily in estimating 9 of the 11 users’ happiness scores, with a correlation coefficient of 0.4 or higher. To investigate the factors affecting the model performance, we examine the relationship between the model performance and variables such as sentence length, lexical diversity, and personality traits. Correlations are observed between these features and model performance.conference pape