# Detecció d'al·lucinacions

Mètodes i tècniques per detectar errors factuals i al·lucinacions en sortides de LLM

Una entrada [Veure totes les etiquetes](https://beancount.io/ca/bean-labs/research-logs/tags) [Retorna a Totes les Publicacions](https://beancount.io/ca/bean-labs/research-logs)

9 de juliol del 2026 7 minuts de lectura

## [Confiança i calibratge en LLM: una enquesta sobre el que realment mostra la recerca](https://beancount.io/ca/bean-labs/research-logs/2026/07/09/confidence-estimation-calibration-llms-survey)

La confiança verbalitzada de GPT-4 només arriba a \~62,7% d'AUROC, poc per sobre de l'atzar. Els agents financers necessiten un millor calibratge.

Mike Thrift

llm ai machine-learning

[Comença gratis→](https://beancount.io/sign-up/?next=%2Fledger%2F) [Documentació](https://beancount.io/ca/docs)

## Where to look next

- [Sitemap](https://beancount.io/sitemap.xml) - all pages with hreflang alternates
- [LLMs index](https://beancount.io/llms.txt) - curated links for AI agents (also /llms-full.txt for full docs)
- [Documentation](https://beancount.io/docs) - every doc is also available as text/markdown via .md suffix or Accept negotiation
- [REST API reference](https://beancount.io/docs/api-reference) - interactive endpoints, schemas, authentication, and code samples
- [OpenAPI spec](https://beancount.io/openapi.json) - OpenAPI 3.1 spec for this website's public content endpoints; the product contract for ledger operations is at https://beancount.io/api-gateway/v1/openapi.json
- [GraphQL](https://beancount.io/graphql) - public, read-only content API over the blog and research logs (no credentials, no ledger data): 12 query fields over 23 types
- [MCP server](https://beancount.io/mcp) - Model Context Protocol at https://beancount.io/api-gateway/mcp
