Only human-human pairs improved accuracy and efficiency across repeated referential rounds, while LVLM-involved pairs stayed verbose and failed to entrain on conceptual pacts.
speech acts
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LVLMs and Humans Ground Differently in Referential Communication
Only human-human pairs improved accuracy and efficiency across repeated referential rounds, while LVLM-involved pairs stayed verbose and failed to entrain on conceptual pacts.