In multi-model self-consuming loops, cross-influences can dampen or invert the alignment benefits of human curation unlike in isolated single-model cases.
10 When and How Human Curation Backfires: Preference Alignment under Multi-Model Self-Consuming Loop Gao, W
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
When and How Human Curation Backfires: Preference Alignment under Multi-Model Self-Consuming Loop
In multi-model self-consuming loops, cross-influences can dampen or invert the alignment benefits of human curation unlike in isolated single-model cases.