Introduces DramaSR-532K benchmark and DramaSR-LRM method using large reasoning models for speaker recognition in long-form TV dramas, claiming outperformance over baselines especially on short utterances.
Merge duplicated dialogue only if it is described twice for the same clip (not if the character actually spoke twice)
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
-
Reasoning LLM Improves Speaker Recognition in Long-form TV Dramas
Introduces DramaSR-532K benchmark and DramaSR-LRM method using large reasoning models for speaker recognition in long-form TV dramas, claiming outperformance over baselines especially on short utterances.