Skip to content

Course Policies

This page provides a technical-course policy template. Replace or augment it with the official university rules.

Collaboration

Discussion is encouraged, but submitted individual work must make each student's contribution clear. Team projects must include a contribution statement and repository history.

Reuse and attribution

Patterns, code fragments, datasets, benchmark suites, and research implementations may be reused when their licenses permit it. Sources must be attributed in code, reports, and presentations.

Generative AI and coding assistants

AI tools can be valuable objects of study in this course. Their use should remain auditable:

  • disclose material AI assistance in assessed work;
  • preserve prompts or interaction logs when AI behavior is part of the experiment;
  • verify generated code with tests and static/dynamic analysis;
  • do not present generated explanations or citations as verified facts without checking them;
  • for research evaluation, separate model assistance used to build the experiment from the model being evaluated.

Reproducibility

Automation-oriented submissions should include enough information to rerun the experiment: environment, versions, inputs, commands, seeds where applicable, raw or derived results, and analysis scripts.

Responsible benchmarking

Do not intentionally contaminate public benchmarks, upload proprietary code to external services, or expose credentials/secrets. Check data and model licenses before redistribution.