LLMs under one-shot greedy decoding enumerate or materialize acceptable sets much worse than they judge membership, a gap that persists across scale, family, and generation and is dominated by omissions.
CodeT: Code generation with generated tests
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Judging Is Not Enumerating: Silent Omissions in LLM-Authored Acceptable Sets
LLMs under one-shot greedy decoding enumerate or materialize acceptable sets much worse than they judge membership, a gap that persists across scale, family, and generation and is dominated by omissions.