We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:


References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Evaluating Predictive Uncertainty under Distributional Shift on Dialogue Dataset

Abstract: In open-domain dialogues, predictive uncertainties are mainly evaluated in a domain shift setting to cope with out-of-distribution inputs. However, in real-world conversations, there could be more extensive distributional shifted inputs than the out-of-distribution. To evaluate this, we first propose two methods, Unknown Word (UW) and Insufficient Context (IC), enabling gradual distributional shifts by corruption on the dialogue dataset. We then investigate the effect of distributional shifts on accuracy and calibration. Our experiments show that the performance of existing uncertainty estimation methods consistently degrades with intensifying the shift. The results suggest that the proposed methods could be useful for evaluating the calibration of dialogue systems under distributional shifts.
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2109.00186 [cs.CL]
  (or arXiv:2109.00186v1 [cs.CL] for this version)

Submission history

From: Nyoungwoo Lee [view email]
[v1] Wed, 1 Sep 2021 04:55:43 GMT (133kb,D)

Link back to: arXiv, form interface, contact.