Q003-S03 — Hallucination Taxonomy — Search Log
Summary
| Field |
Value |
| Source/Database |
Web search (targeted: academic taxonomy, NIST) |
| Query terms |
Multiple: "arxiv hallucination spectrum taxonomy LLM types severity"; "NIST AI risk management framework confabulation" |
| Filters |
None |
| Results returned |
20 (across 2 searches) |
| Results selected |
2 |
| Results rejected |
18 |
Selected Results
| Result |
Title |
URL |
Rationale |
| S03-R01 |
A Survey on Hallucination in LLMs |
https://arxiv.org/abs/2311.05232 |
Comprehensive academic taxonomy |
| S03-R02 |
NIST AI 600-1 |
https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf |
Official US framework for confabulation |
Rejected Results
| Result |
Title |
URL |
Rationale |
| (18 results) |
Various |
Various |
Duplicate surveys, tangential topics, or less comprehensive than selected |
Notes
Academic hallucination research is extensive but focused on detection and mitigation techniques rather than on how to communicate the spectrum to end users. The gap between technical understanding and user-facing explanation is a core finding.