BERTTune: Fine-Tuning Neural Machine Translation with BERTScore

Unanue, Iñigo Jauregi; Parnell, J. Jacob; Piccardi, Massimo

doi:10.48550/arxiv.2106.02208

Cited by 3 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given the non-uniqueness of the output content from LLM-based embodied model, a similarity calculation approach is employed to assess the alignment with the target task. Common methods for calculating this similarity include those based on the bag-of-words model [35], TF-IDF weighted word vectors [1], Bert-score [30], among others. These methods are predominantly utilized in machine translation and text matching tasks.…”

Section: 23mentioning

confidence: 99%

Geography of Technology Transfer in China

Liu

2023

East China Normal University Scientific Reports

View full text Add to dashboard Cite

Graphs can model complex relationships between objects, enabling a myriad of Web applications such as online page/article classification and social recommendation. While graph neural networks (GNNs) have emerged as a powerful tool for graph representation learning, in an end-to-end supervised setting, their performance heavily relies on a large amount of task-specific supervision. To reduce labeling requirement, the "pre-train, fine-tune" and "pre-train, prompt" paradigms have become increasingly common. In particular, prompting is a popular alternative to fine-tuning in natural language processing, which is designed to narrow the gap between pre-training and downstream objectives in a task-specific manner. However, existing study of prompting on graphs is still limited, lacking a universal treatment to appeal to different downstream tasks. In this paper, we propose GraphPrompt, a novel pre-training and prompting framework on graphs. GraphPrompt not only unifies pre-training and downstream tasks into a common task template, but also employs a learnable prompt to assist a downstream task in locating the most relevant knowledge from the pre-trained model in a task-specific manner. Finally, we conduct extensive experiments on five public datasets to evaluate and analyze GraphPrompt. CCS CONCEPTS• Computing methodologies → Learning latent representations; • Information systems → Data mining.

show abstract

Section: 23mentioning

confidence: 99%

Geography of Technology Transfer in China

Liu

2023

East China Normal University Scientific Reports

View full text Add to dashboard Cite

show abstract

“…As a further advancement in caption quality assessment, researchers are exploring learning-based evaluation strategies [155]- [160]. This approach employs a component within a complete captioning system responsible for assessing the VOLUME 4, 2016 This article has been accepted for publication in IEEE Access.…”

Section: ) Evaluation Through Learning-based Methodsmentioning

confidence: 99%

Deep Learning Approaches for Image Captioning: Opportunities, Challenges and Future Potential

Jamil,

Saif-Ur-Rehman,

Mahmood

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Generative intelligence relies heavily on the integration of vision and language. Much of the research has focused on image captioning, which involves describing images with meaningful sentences. Typically, when generating sentences that describe the visual content, a language model and a vision encoder are commonly employed. Because of the incorporation of object areas, properties, multi-modal connections, attentive techniques, and early fusion approaches like bidirectional encoder representations from transformers (BERT), these components have experienced substantial advancements over the years. This research offers a reference to the body of literature, identifies emerging trends in an area that blends computer vision as well as natural language processing in order to maximize their complementary effects, and identifies the most significant technological improvements in architectures employed for image captioning. It also discusses various problem variants and open challenges. This comparison allows for an objective assessment of different techniques, architectures, and training strategies by identifying the most significant technical innovations, and offers valuable insights into the current landscape of image captioning research.

show abstract

“…BLEU, ROUGE etc). BERTScore has been used as an evaluation metric for image captioning [40], summarization ( [21]), machine translation ( [37]) etc. BERTScore returns a score (0 − 1) between student and reference cause/effect.…”

Section: Cause-effect Error Classifiermentioning

confidence: 99%

Few-Shot Question Generation for Personalized Feedback in Intelligent Tutoring Systems

Kulshreshtha

Shayan²,

Belfer³

et al. 2022

Pais 2022

View full text Add to dashboard Cite

Existing work on generating hints in Intelligent Tutoring Systems (ITS) focuses mostly on manual and non-personalized feedback. In this work, we explore automatically generated questions as personalized feedback in an ITS. Our personalized feedback can pinpoint correct and incorrect or missing phrases in student answers as well as guide them towards correct answer by asking a question in natural language. Our approach combines cause–effect analysis to break down student answers using text similarity-based NLP Transformer models to identify correct and incorrect or missing parts. We train a few-shot Neural Question Generation and Question Re-ranking models to show questions addressing components missing in the student’s answers which steers students towards the correct answer. Our model vastly outperforms both simple and strong baselines in terms of student learning gains by 45% and 35% respectively when tested in a real dialogue-based ITS. Finally, we show that our personalized corrective feedback system has the potential to improve Generative Question Answering systems.

show abstract

BERTTune: Fine-Tuning Neural Machine Translation with BERTScore

Cited by 3 publications

References 10 publications

Geography of Technology Transfer in China

Geography of Technology Transfer in China

Deep Learning Approaches for Image Captioning: Opportunities, Challenges and Future Potential

Few-Shot Question Generation for Personalized Feedback in Intelligent Tutoring Systems

Contact Info

Product

Resources

About