LLMs fail at context-conditioned next-utterance anticipation in multimodal dialogues; SayNext-Bench and SayNext-Chat demonstrate the value of explicit multimodal and anticipatory processing.
Title resolution pending
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
-
SayNext-Bench: Why Do LLMs Struggle with Next-Utterance Anticipation?
LLMs fail at context-conditioned next-utterance anticipation in multimodal dialogues; SayNext-Bench and SayNext-Chat demonstrate the value of explicit multimodal and anticipatory processing.