Skip to content

Evaluation and grading tasks: scoring model outputs against a rubric

Pathwize EditorialExpert data and AI evaluation2 min read
GuidesPathwize AI

When you need a number you can track, you need an evaluation task. How rubrics turn a vague sense of quality into a measurement graders agree on.

When you need a number you can track across releases, you need an evaluation task. An expert scores a single model output against a rubric, either pass/fail or on a scale. Unlike a preference task, there is no second answer to compare against: the standard is the rubric, and the job is to apply it consistently.

The rubric is the whole game

An evaluation is only as good as its rubric. A vague instruction (rate the quality from 1 to 5) produces five graders and five private definitions of quality. A good rubric names the dimensions, defines each level with an example, and tells the grader how to handle partial credit and edge cases. The aim is that two qualified experts, grading the same output, land in the same place.

Pass/fail or a scale?

Pass/fail is best for gates: does this output meet the bar to ship, yes or no? It is fast, hard to game, and easy to act on. A scale (say 1 to 5, or a set of labelled bands) carries more information and is better for tracking gradual improvement, but it demands a tighter rubric so the bands mean the same thing to everyone. Pick the one that matches the decision you will make with the result.

Measuring the graders, not just the model

Because evaluation produces a headline number, it is worth checking that the number is trustworthy. Send some items to more than one grader and watch inter-rater agreement. High agreement means your rubric is doing its job. Low agreement is a signal that the rubric is ambiguous, not that your experts are careless, and it is far cheaper to catch before the number reaches a slide.

Where evaluation fits

Evaluation is how you benchmark a release, run a regression check, and set a quality gate. Preference tells you which of two answers is better; evaluation tells you how good a single answer is against a fixed standard. In a data programme, evaluation is the measurement stage: the part that tells you whether the fine-tuning and preference work actually moved the model.

On Pathwize

An evaluation task on Pathwize sends each output to credential-verified experts, records the score and the rubric it was graded against, and keeps a signed trail of who graded what. You get a benchmark you can defend and reproduce, not a number whose origin you cannot reconstruct. Create one from the AI tasks console and set how many experts grade each item.

Frequently asked questions

What is an evaluation or grading task?+

A task where an expert scores a single model output against a rubric, either pass/fail or on a scale. It produces a measurement you can track, unlike a preference task, which compares two outputs.

Should I use pass/fail or a scale?+

Use pass/fail for gates and quick ship/no-ship decisions. Use a scale to track gradual improvement, but write a tighter rubric so the bands mean the same thing to every grader. Match the format to the decision you will make.

How do I know my evaluation scores are reliable?+

Send some items to more than one grader and watch inter-rater agreement. High agreement means the rubric is clear; low agreement usually means the rubric is ambiguous and needs tightening.

Related reading

See Pathwize on your own data

Source verifiable expert data with provenance built in, EU-native and audit-ready.

Book a demo
← All posts

Related stories

GuidesPathwize AI

How to evaluate an AI data vendor: a due-diligence checklist

The questions that separate a data vendor you can defend to an auditor from one that will cost you a re-labelling project six months in.

Pathwize AIGuides

How to run a human-data pilot before you commit

A pilot is the cheapest way to learn whether a data partner can actually do your hardest work. Here is how to design one that tells you something real.

GuidesPathwize AI

The four types of AI data tasks, explained

Preference, evaluation, red teaming and domain judgment. What each task type does, when to use it, and how they fit together in a data programme.