Knowing When Not to Answer: Cross-Domain and Multi-Turn Generalization of Latent Underspecification Signals

Excerpt

arXiv:2610.08413v1 Announce Type: cross Abstract: Large language models routinely answer questions that cannot be answered from the information given, and in dialogue they answer before enough has been said. Unanswerability is linearly decodable from hidden states, but it is unclear which of its forms share a representation and whether the signal is useful in dialogue. We contribute a turn-labeled multi-turn benchmark (423 conversations, 1,661 labeled turn-states) and an evaluation harness with