Worst-case seeker simulations show that emotional support dialogue systems suffer substantial performance drops, with large general LLMs more robust than specialized models but still limited in sustaining engagement.
Preprint, arXiv:2601.01530
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
When Seekers Are Hard to Help: Evaluating Emotional Support Dialogue Systems in Worst-Case Interactions
Worst-case seeker simulations show that emotional support dialogue systems suffer substantial performance drops, with large general LLMs more robust than specialized models but still limited in sustaining engagement.