Índice alfabético
Índice alfabético de los términos y conceptos clave del Manual QA AI v13 (Apéndice C). Para definiciones operativas, ver Conceptos clave. Para definiciones canónicas, ver el Cap. 33 del manual.
Cómo buscar
Si conoces el nombre del concepto, este índice te lleva al capítulo. Si conoces solo el tema, empieza por la página de Conceptos que está ordenada por área temática.
A
| Término | Sección |
|---|
| Agentes AI | Cap. 21 |
| Aggregation bias | §25.1 |
| Alucinaciones | Cap. 17 |
| Answer Correctness | §7.1, §7.4 |
| Answer Relevancy | §7.1, §7.4 |
| Antipatrones de evaluación | Cap. 22 |
| Antipatrones operativos | Cap. 26 |
| Argilla | §31.5 |
| AutoGen | §21.1 |
B
| Término | Sección |
|---|
| BERTScore | §11.2 |
| Bias (sesgo) | Cap. 25 |
| BM25 + hybrid search | §29.2 |
C
| Término | Sección |
|---|
| Canary deployment | §18.2 |
| Canary tokens | §28.3 |
| Chunking | §6.2 |
| CI/CD quality gates | Cap. 18 |
| Citation requirement | §17.3 |
| Coherencia multi-turno | Cap. 16 |
| Confidence calibration | §17.3 |
| Consistency | §16.3, §12.3 |
| Consistency score (robustness) | §12.3 |
| Context Precision | §7.1 |
| Context Recall | §7.1 |
| Coreference resolution | §16.3 |
| Cosine similarity | §11.2 |
| Cost-aware QA | Cap. 27 |
| Cost regression testing | §27.4 |
| CrewAI | §21.1 |
| Cross-encoder | §6.1 |
D
| Término | Sección |
|---|
| DeepEval | §20.3 |
| Deployment bias | §25.1 |
| Detoxify | §25.3 |
| Drift (deriva semántica) | Cap. 13 |
| Drift detector | §13.3 |
E
| Término | Sección |
|---|
| Embedding | §6.1 |
| EU AI Act | §3.7 |
| Evaluation bias | §25.1 |
| Excessive Agency (LLM06) | Cap. 14 |
| Execution trace | §21.4 |
| Extrinsic hallucination | §17.1 |
F
| Término | Sección |
|---|
| Faithfulness | §7.1, §7.3 |
| Faithfulness vs. factual accuracy | §7.3 |
| False refusal rate | §25.4 |
| Few-shot examples | §8.3 |
| FormatChecker (jsonschema) | §30.3 |
| Function calling | Cap. 30 |
G
| Término | Sección |
|---|
| Golden datasets | Cap. 9 |
| Golden traces | §21.7 |
| Ground truth | §7.1, §9.3 |
| Groundedness | Cap. 33 |
H
| Término | Sección |
|---|
| Hallucinations | Cap. 17 |
| Historical bias | §25.1 |
| Human-in-the-loop | Cap. 31 |
| HyDE | §29.2 |
| Hybrid search | §29.2 |
I
| Término | Sección |
|---|
| Improper Output Handling (LLM05) | Cap. 14 |
| Indirect prompt injection | §15.1 |
| Indexación | §6.1 |
| Information Retrieval (métricas) | §19.3 |
| Inter-annotator agreement | Cap. 31 |
| Intrinsic hallucination | §17.1 |
| ISO 25010 | §3.6 |
| ISO/IEC 42001 | §3.7 |
J
| Término | Sección |
|---|
| Jailbreak | §15.1 |
| JSON Schema validation | §30.3 |
K
| Término | Sección |
|---|
| κ de Cohen | §31.2 |
| Knowledge drift | §13.1 |
| Krippendorff α | §31.2 |
| KS test (Kolmogorov-Smirnov) | §13.3 |
L
| Término | Sección |
|---|
| Label Studio | §31.5 |
| Langfuse | §20.1 |
| Latencia P95 / P99 | §27.2 |
| LlamaGuard | §25.3 |
| LLM-as-Judge | Cap. 8 |
| LLM puro | §5.1 |
| L1–L5 madurez | Cap. 23 |
M
| Término | Sección |
|---|
| MAP@k | §19.3 |
| Matriz de calibración de umbrales | §4.3 |
| Measurement bias | §25.1 |
| Memory window (multi-turno) | §16.3 |
| Misinformation (LLM09) | Cap. 14 |
| Modelo de madurez QA AI | §23.1 |
| MRR@k | §19.3 |
| Multi-agente | §21.9 |
| Multi-query retrieval | §29.2 |
| Multi-turno | Cap. 16 |
N
| Término | Sección |
|---|
| NDCG@k | §19.3 |
| NeMo Guardrails | §25.3 |
| NIST AI RMF 1.0 | §3.7 |
O
| Término | Sección |
|---|
| Observabilidad | Cap. 19 |
| OpenAI Moderation | §25.3 |
| OWASP LLM Top 10 | Cap. 14 |
P
| Término | Sección |
|---|
| Parent-child chunks | §29.2 |
| Payload splitting | §15.1 |
| Permission boundary | §21.6 |
| Perspective API | §25.3 |
| Perturbation testing | Cap. 12 |
| PII (Personally Identifiable Info) | Cap. 28 |
| PII scrubbing | §28.4 |
| Plan quality score | §21.3 |
| Position bias | §8.2 |
| Privacy testing | Cap. 28 |
| PromptBench | §20.1, Cap. 12 |
| Prompt caching | §27.5 |
| Prompt Injection (LLM01) | Cap. 15 |
| Prompt leaking | §15.1, §28.3 |
| Prompt Regression Testing | Cap. 24 |
| PSI (Population Stability Index) | Cap. 13 |
Q
| Término | Sección |
|---|
| Quality gate | Cap. 18 |
| Query rewriting | §29.2 |
| Query-side drift | §13.1 |
R
| Término | Sección |
|---|
| RAG (Retrieval-Augmented Generation) | Cap. 6 |
| RAGAS | Cap. 7 |
| ranx (librería) | §19.3 |
| Refusal rate | §25.4 |
| Regression testing | §10.1, Cap. 24, §27.4 |
| Representation bias | §25.1 |
| Reranker | §6.1, §29.2 |
| Reproducibilidad y determinismo | Cap. 32 |
| Response-side drift | §13.1 |
| Retrieval avanzado | Cap. 29 |
| RiskLevel | §4.3 |
| Robustness | Cap. 12 |
S
| Término | Sección |
|---|
| Safety testing | §25.4 |
| Schema mismatch | §30.2 |
| Self-enhancement bias | §8.2 |
| Self-RAG | §29.2 |
| Semantic chunking | §6.2 |
| Sentence-window retrieval | §29.2 |
| Sensitive Information Disclosure (LLM02) | Cap. 14 |
| Shadow AI | §3.8 |
| Shadow traffic | §18.2 |
| Similitud semántica | Cap. 11 |
| Smolagents | §21.1 |
| Supply Chain (LLM03) | Cap. 14 |
| Synthetic data | §9.4 |
| System prompt hardening | §15.3 |
| System Prompt Leakage (LLM07) | Cap. 14 |
T
| Término | Sección |
|---|
| Tabla maestra de umbrales | §4.3, §18.2 |
| Temporal hallucination | §17.1 |
| Time-to-first-token | §27.2 |
| Tool fan-out | §27.2 |
| Tool selection accuracy | §21.7 |
| Tool use testing | Cap. 30 |
| Topic tracking | §16.3 |
| Toxicity testing | §25.3 |
| Trace schema | §19.2 |
| TruLens | §20.2 |
U
| Término | Sección |
|---|
| Umbrales por dominio | §4.3, §18.2 |
| Unbounded Consumption (LLM10) | Cap. 14 |
V
| Término | Sección |
|---|
| Vector and Embedding Weaknesses (LLM08) | Cap. 14 |
| Vector store | §6.1, Cap. 33 |
| Verbosity bias | §8.2 |
Y
| Término | Sección |
|---|
| YAML quality gate (GitHub Actions) | §18.3 |
Δ
| Término | Sección |
|---|
| Δ (delta) — variación de métrica | Cap. 33, Tabla 4.2 |
