Prueba sobre métodos de detección de alucinaciones

Las respuestas y explicaciones aparecen a medida que realiza el cuestionario.

1. ¿Qué observación subyace a los métodos de autoconsistencia como SelfCheckGPT?
   • When a model knows a fact, repeated samples tend to agree, while fabricated details tend to vary
   • Models always give identical answers to repeated questions
   • Longer answers are always more accurate
   • Setting temperature to zero removes all hallucinations
2. ¿Cuál es una limitación clave de la verificación de la autoconsistencia?
   • It requires a large labeled dataset before use
   • It works only on image outputs
   • It misses errors the model repeats consistently, and it costs extra model calls
   • It cannot run without retrieved source documents
3. ¿Qué miden realmente los controles de vinculación basados ​​en la fuente?
   • The model's training cutoff date
   • Faithfulness to the provided sources, which is not the same as truth if the sources are wrong
   • How quickly the model responds
   • Grammar and spelling quality
4. ¿En qué se diferencia la entropía semántica de la simple incertidumbre a nivel de token?
   • It counts the number of words in each answer
   • It relies only on human raters
   • It avoids sampling multiple answers
   • It groups sampled answers by meaning before measuring uncertainty
5. ¿Por qué la guía recomienda exigir al verificador que cite el tramo de soporte?
   • Hace que la salida sea más corta.
   • It makes verification auditable, and a simple string match confirms the quote appears in the source
   • It makes the verifier more creative
   • It reduces verification cost to zero
6. Which checks should run first in a detection pipeline?
   • Cheap deterministic checks, such as confirming that cited URLs, IDs or case numbers exist
   • Full human review of every answer
   • Fine-tuning the model on the flagged answers
   • Semantic entropy on every token
7. What caveat applies to using an LLM as a judge?
   • Judges cannot read source documents
   • Judges work only in English
   • Judges are always correct
   • Judges have their own biases and error rates, so they should be checked against human labels
8. Why is simply asking a model 'are you sure?' an unreliable detection method?
   • Models cannot read follow-up questions
   • It is prohibited by API terms
   • Models may abandon correct answers or defend wrong ones
   • Duplica el tamaño de la ventana de contexto.