Datos con muchos empates
A large tie class can make aggregate agreement misleading.
Cálculo / simulación localExperimento educativo local, sin API externa. Los datos sintéticos no son mediciones de producción.
Study guide & lesson outline
LEARNING CONTRACTUnderstand → implement → investigate
Goal: explain the mechanism in this lesson, reproduce one example and test a failure case. Reading alone does not mark mastery.
- Understand: state assumptions, shapes/units and a counterexample.
- Implement: reproduce a numerical example and inspect boundaries.
- Investigate: compare one alternative under a matched budget.
02 · Predice y configura
Las notas técnicas, los parámetros y las respuestas están actualmente en inglés.
03 · Analiza la evidencia
Métodos y referencias
Build an exportable rubric and analyze supplied judge labels. No LLM is called. In swapped trials A means original B and vice versa; normalize before computing consistency. Fixture labels are synthetic. Rubric wording is an original starting point, not a validated scale.
Text / 原始公式
agreement=matches/n; consistency=matches(original, remapped swapped)/n
High agreement on a small biased sample does not validate a judge for a new domain.
Judging LLM-as-a-Judge · MT-Bench ↗Preguntas de práctica
judge-4 / 01What does “Tie-heavy dataset” test?
A large tie class can make aggregate agreement misleading.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 02Which computation produces the result in “Tie-heavy dataset”?
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 03What is the baseline configuration for “Tie-heavy dataset”?
{"task":"Compare two answers for factual correctness using supplied evidence.","ratings":"[{\"human\":\"tie\",\"judge\":\"tie\",\"swapped\":\"tie\"},{\"human\":\"tie\",\"judge\":\"A\",\"swapped\":\"B\"},{\"human\":\"tie\",\"judge\":\"tie\",\"swapped\":\"tie\"}]"}
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 04What numerical result should you check for “Tie-heavy dataset”?
{"Human agreement":0.6666666666666666,"Order consistency":1,"A wins (original)":0.3333333333333333}
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 05How would you run a controlled comparison for “Tie-heavy dataset”?
Keep the supplied fixture and other settings fixed; change ratings. Compare human agreement and order consistency; retain evidence spans in real evaluations.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 06What failure could invalidate a conclusion from “Tie-heavy dataset”?
High agreement on a small biased sample does not validate a judge for a new domain.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 07Is “Tie-heavy dataset” a benchmark of a deployed language model?
No. Build an exportable rubric and analyze supplied judge labels. No LLM is called. In swapped trials A means original B and vice versa; normalize before computing consistency. Fixture labels are synthetic. Rubric wording is an original starting point, not a validated scale.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 08Which evidence should you report for “Tie-heavy dataset”?
Compare human agreement and order consistency; retain evidence spans in real evaluations. Export inputs, engine, result and method; distinguish synthetic fixtures from observed production data.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 09How can another researcher reproduce “Tie-heavy dataset”?
Open judge-4, reset the preset, retain the exact inputs and run. The calculation is deterministic for identical inputs.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →judge-4 / 10What is the next validation step after “Tie-heavy dataset”?
High agreement on a small biased sample does not validate a judge for a new domain. Test the hypothesis on a representative held-out workload and document a baseline, success metric and stopping rule before claiming deployment readiness.
Original teaching answer / local derivation. Method reference: Judging LLM-as-a-Judge · MT-Bench ↗
Abrir laboratorio →Assumptions, implementation & acceptance
Comparison count must be positive. Preserve ties, map swapped labels back to original identities, and separately report parse failures. Synthetic ratings do not establish a real judge’s reliability.
Engine formulas, scenario presets and numerical safeguards reviewed. Seeded synthetic inputs are not a production benchmark.
Before running: predict a boundary case
A large tie class can make aggregate agreement misleading.
Hint: isolate the assumption
Keep the baseline input fixed, change one parameter or boundary, and compare both the numerical result and the claimed units. A passing example verifies this fixture only.
Minimal executable example
Save the example and linked modules in one folder. Use Node.js 22 or newer, then run node judge-4.mjs. No third-party runtime dependencies.
import {calculate,network} from './academy-core.mjs';
const params={
"task": "Compare two answers for factual correctness using supplied evidence.",
"ratings": "[{\"human\":\"tie\",\"judge\":\"tie\",\"swapped\":\"tie\"},{\"human\":\"tie\",\"judge\":\"A\",\"swapped\":\"B\"},{\"human\":\"tie\",\"judge\":\"tie\",\"swapped\":\"tie\"}]"
};
const result=calculate('judge',params);
console.log(result.metrics);Test and transfer
A large tie class can make aggregate agreement misleading.
Record input, code version, expected result and observed result in the notebook below. Then construct a counterexample that violates the assumption above and explain why the original conclusion no longer follows.
Download complete kit (.tar.gz) · Complete reference kit and regression checksPredict, test and keep evidence
Before revealing an answer, predict one result. After the exercise, record the discrepancy and a transfer test. Saved locally across language editions.
Open notebook / export / clear →The mathematical review scope is recorded per lesson. It does not establish empirical paper reproduction; report suspected errors in your notes. Review scope