Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Training an NLP Scholar at a Small Liberal Arts College: A Backwards Designed Course Proposal

Grusha Prasad, Forrest Davis

TL;DR

The paper argues that at Small Liberal Arts Colleges it is more feasible to train NLP scholars than engineers due to curriculum and student background constraints. It presents a backwards designed upper-level course anchored by a capstone project, defining three core scholar skills (describe language processing, effectively use existing tools, and evaluate NLP claims) and six tenets (complexity, multilingualism, abstraction, tool use, benchmarks, and hype critique). The course emphasizes a layered introduction to concepts, a modular toolkit, lab-lecture integration, a midterm replication, and a final group project with posters and individual papers, plus a society reflection to connect NLP practice with societal implications. The framework is discussed for adaptability to other institutions and disciplines, highlighting the value of the NLP scholar paradigm for curricular design and broader educational impact in NLP.

Abstract

The rapid growth in natural language processing (NLP) over the last couple years has generated student interest and excitement in learning more about the field. In this paper, we present two types of students that NLP courses might want to train. First, an "NLP engineer" who is able to flexibly design, build and apply new technologies in NLP for a wide range of tasks. Second, an "NLP scholar" who is able to pose, refine and answer questions in NLP and how it relates to the society, while also learning to effectively communicate these answers to a broader audience. While these two types of skills are not mutually exclusive -- NLP engineers should be able to think critically, and NLP scholars should be able to build systems -- we think that courses can differ in the balance of these skills. As educators at Small Liberal Arts Colleges, the strengths of our students and our institution favors an approach that is better suited to train NLP scholars. In this paper we articulate what kinds of skills an NLP scholar should have, and then adopt a backwards design to propose course components that can aid the acquisition of these skills.

Training an NLP Scholar at a Small Liberal Arts College: A Backwards Designed Course Proposal

TL;DR

The paper argues that at Small Liberal Arts Colleges it is more feasible to train NLP scholars than engineers due to curriculum and student background constraints. It presents a backwards designed upper-level course anchored by a capstone project, defining three core scholar skills (describe language processing, effectively use existing tools, and evaluate NLP claims) and six tenets (complexity, multilingualism, abstraction, tool use, benchmarks, and hype critique). The course emphasizes a layered introduction to concepts, a modular toolkit, lab-lecture integration, a midterm replication, and a final group project with posters and individual papers, plus a society reflection to connect NLP practice with societal implications. The framework is discussed for adaptability to other institutions and disciplines, highlighting the value of the NLP scholar paradigm for curricular design and broader educational impact in NLP.

Abstract

The rapid growth in natural language processing (NLP) over the last couple years has generated student interest and excitement in learning more about the field. In this paper, we present two types of students that NLP courses might want to train. First, an "NLP engineer" who is able to flexibly design, build and apply new technologies in NLP for a wide range of tasks. Second, an "NLP scholar" who is able to pose, refine and answer questions in NLP and how it relates to the society, while also learning to effectively communicate these answers to a broader audience. While these two types of skills are not mutually exclusive -- NLP engineers should be able to think critically, and NLP scholars should be able to build systems -- we think that courses can differ in the balance of these skills. As educators at Small Liberal Arts Colleges, the strengths of our students and our institution favors an approach that is better suited to train NLP scholars. In this paper we articulate what kinds of skills an NLP scholar should have, and then adopt a backwards design to propose course components that can aid the acquisition of these skills.
Paper Structure (57 sections, 2 figures, 3 tables)

This paper contains 57 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Teaching NLP as an upper level course at a Small Liberal Arts College
  3. Required and elective courses at Colgate
  4. Take-aways from previous iterations of teaching NLP and Applied ML
  5. Order of introducing non-neural vs. neural approaches
  6. Prior programming experience
  7. How to train an NLP Scholar
  8. S1: Describe language processing
  9. S2: Effectively using existing NLP tools
  10. S3: Evaluate claims about NLP systems
  11. Tenets of our proposed course
  12. T1: Appreciate the complexity of language
  13. T2: Focus on multilingualism
  14. T3: Recognize how NLP tasks abstract away from the complexity
  15. T4: Build NLP systems using existing tools
  16. ...and 42 more sections

Figures (2)

  • Figure 1: Sketch of the core components of the toolkit
  • Figure 2: Final poster template