Skip to content

Exams, Papers, and Projects: Three Forms of Learning Evidence | CMU 10-423 Topic Guide

Sep 30, 20261 min
Home›ai
TL;DRAn exam tests reasoning under a fixed question. A paper report tests your judgment of someone else’s evidence. A project tests your ability to produce reproducible evidence. These are distinct learning tasks, now taught separately while this URL remains an assessment map.
#cmu-10423#ai-course#cmu#exam-prep#paper-reading#research-project#self-study
Series: Reading CMU 10-423 Generative AI (46 / 47)
Audio, Video, and World Models: A Reading Map | CMU 10-423 Topic Guide

Look up & save

Loading dictionary…

🌏 中文版

Video status: recordings require login or course authorization. Recording source

An exam tests reasoning under a fixed question. A paper report tests your judgment of someone else’s evidence. A project tests your ability to produce reproducible evidence. These are distinct learning tasks, now taught separately while this URL remains an assessment map.

Choose a lesson

  • Practice exam: work through its mechanism, example, and answer criteria.
  • HW623 paper report: work through its mechanism, example, and answer criteria.
  • Final project: work through its mechanism, example, and answer criteria.

The course map lists the core path. This page is a topic guide, not an additional lecture. Choose the question you need to answer and follow its lesson.

A worked comparison

Teaching example: computing a lower average loss for model A tests arithmetic. Reading a paper additionally requires tracing data, baselines, and conditions. A project must retain code, data versions, seeds, and outcomes. If A is tested on 100 easy cases and B on 100 hard ones, averages do not establish that A is the better model. Explain the problem in an exam, identify it in a report, and run a shared-test-set comparison in a project. Each step produces different evidence.

Check your understanding

Do working code, fluent writing, or a correct exam option independently validate a research conclusion?

Answer and criterion

No. Code needs sound evaluation, writing needs evidence and attribution, and an answer needs reasoning that survives a variation. The three lessons provide separate criteria.

How to use this map

Before reading the answer, identify the inputs, what changes, and the outputs. Then complete the linked lesson’s variation. If two mechanisms seem identical, trace where their conditions or data paths diverge. This page checks relationships and boundaries; it does not replace the derivations, implementations, or evaluations in the individual lessons.

The source edition is Spring 2026. Teaching examples are not official assessments or reconstructed classroom speech. Recordings remain subject to course authorization.

Course video sources

The official Spring 2026 course site directs recordings to SCS Panopto. On 2026-10-12, the anonymous reader returned an unloaded folder and a sign-in prompt; it did not provide playable lecture content. This establishes an access limit, not the absence of recordings. This guide uses public written materials and provides no reconstructed classroom speech or original-audio transcript.

Update Log

  • 2026-10-11: Rebuilt the teaching scope, worked example, and self-check; retained the original URL and publication date.

  • 2026-10-10: Added explicit video status and checked recording sources and access notes.

  • 2026-10-10: Rechecked video status. The login wall is confirmed (anonymous Panopto folder lists no videos and prompts sign-in); the official pages link no public YouTube version, so the status is unchanged.

References

Related

A Generative AI Learning Path: Reading the 35 Core Guides | CMU 10-423 Course Overview1 minFinal Projects: Build a Reproducible Generative-Model Experiment | CMU 10-423 Topic Guide1 minPaper Reports: Turn a Model Summary into a Testable Argument | CMU 10-423 Assignment 6231 min