Related papers: Newspaper Story Problems and Other Tasks for Conte…
We introduce SemCSE, an unsupervised method for learning semantic embeddings of scientific texts. Building on recent advances in contrastive learning for text embeddings, our approach leverages LLM-generated summaries of scientific…
Over the past year and a half we have developed an innovative approach to the teaching of `Physical Science', a general education course typically found in the curricula of nearly every college and university. The new approach uses popular…
In this paper, we focus on a new practical task, document-scale text content manipulation, which is the opposite of text style transfer and aims to preserve text styles while altering the content. In detail, the input is a set of structured…
Unsupervised sentence embedding aims to obtain the most appropriate embedding for a sentence to reflect its semantic. Contrastive learning has been attracting developing attention. For a sentence, current models utilize diverse data…
Reading comprehension continues to be a crucial research focus in the NLP community. Recent advances in Machine Reading Comprehension (MRC) have mostly centered on literal comprehension, referring to the surface-level understanding of…
Image-text matching plays a central role in bridging vision and language. Most existing approaches only rely on the image-text instance pair to learn their representations, thereby exploiting their matching relationships and making the…
Despite the empirical success of foundation models, we do not have a systematic characterization of the representations that these models learn. In this paper, we establish the contexture theory. It shows that a large class of…
Slide presentations are an effective and efficient tool used by the teaching community for classroom communication. However, this teaching model can be challenging for blind and visually impaired (VI) students. The VI student required…
Continual Learning (CL) sequentially learns new tasks like human beings, with the goal to achieve better Stability (S, remembering past tasks) and Plasticity (P, adapting to new tasks). Due to the fact that past training data is not…
This is a report of a course on modern physics designed and taught to undergraduate science and engineering students in the Spring of 2013. The course, meant for freshmen, attempts to integrate statistical mechanics into non-classical…
A primary goal of many undergraduate- and graduate-level courses in the physical sciences is to prepare students to engage in scientific research, or to prepare students for careers that leverage skillsets similar to those used by research…
The ability of language models to learn a task from a few examples in context has generated substantial interest. Here, we provide a perspective that situates this type of supervised few-shot learning within a much broader spectrum of…
Background: Dataset skills are used in STEM fields from healthcare work to astronomy research. Few fields explicitly teach students the skills to analyze datasets, and yet the increasing push for authentic science implies these skills…
This paper describes the development of a new university physics course designed to integrate physics, education, research, and community partnerships. The coordinated system of activities links the new course to local community efforts in…
Dense document embeddings are central to neural retrieval. The dominant paradigm is to train and construct embeddings by running encoders directly on individual documents. In this work, we argue that these embeddings, while effective, are…
Flipped classroom approach has gained attention for educational practitioners and researchers in recent years. In contrast with traditional classroom, in flipped classroom, students gather basic knowledge out of class, so that class time…
This paper is concerned with the design and implementation of an innovative user support system in the frame of an open educational environment. The environment adapted is ModelsCreator (MC), an educational system supporting learning…
This study discusses the opportunity to integrate tinkering, a constructionist practice, into formal education, highlighting its potential and challenges. We propose a model through which teachers can blend the open exploratory nature of…
Debugging physical computing projects provides a rich context to understand cross-disciplinary problem solving that integrates multiple domains of computing and engineering. Yet understanding and assessing students' learning of debugging…
To better understand how student and faculty perceptions of the learning climate in science/mathematics classes influence success and persistence, we followed a cohort of 1425 academically able students who entered CEGEP in the fall of…