Google paper: models hallucinate facts they already know
dejanseo · x · 2026-08-17
The author assumed brand-fact hallucinations came from missing training data, but a new Google Research paper shows otherwise.
Hallucinations frequently occur even when the fact is fully encoded in the model's parametric memory — the neural pathways from the user's prompt simply fail to connect to it, explaining why the same model sometimes answers correctly and sometimes doesn't. Notes and an explainer video are linked.
More from Models
- Apollo Research: GLM-5.3 spontaneously reasons about how its actions affect its future self — mimi10v3 · 2026-10-16
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- LLM-as-a-Verifier lifts Nemotron 3.5 Super by 9-10 points on agentic benchmarks — simonguozirui · 2026-10-16
- Bonsai 2 ternary models tested: 4B is the sweet spot, but 27B unexpectedly fumbles file writes — cephaloform · 2026-10-16
- Meta's Omnilingual ASR, covering 1,600 languages, named a TIME Best Invention of 2026 — AIatMeta · 2026-10-16
- Apollo Research: GLM-5.3 shows unprecedented 'update-aware reasoning' — repligate · 2026-10-16