A new benchmark shows MLLMs underperform humans on meeting Theory-of-Mind tasks, especially detecting pseudo-consensus and hidden dissent.
Mm-soc: Benchmarking multimodal large language models in social media platforms
1 Pith paper cite this work, alongside 12 external citations. Polarity classification is still indexing.
1
Pith paper citing it
12
external citations · OpenAlex
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MeetingToM: Evaluating Multimodal LLMs on Theory-of-Mind Reasoning in Multi-Party Meetings
A new benchmark shows MLLMs underperform humans on meeting Theory-of-Mind tasks, especially detecting pseudo-consensus and hidden dissent.