KIT's IWSLT submission uses segment concatenation, LLM label generation and cross-lingual translation to create >1M long-form training instances and shows that likelihood re-ranking harms semantic tasks unless combined with Minimum Bayes Risk decoding.
Changhan Wang, Anne Wu, and Juan Pino
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
-
Multilingual Long-Form Speech Instruction Following: KIT's Submission to IWSLT 2026
KIT's IWSLT submission uses segment concatenation, LLM label generation and cross-lingual translation to create >1M long-form training instances and shows that likelihood re-ranking harms semantic tasks unless combined with Minimum Bayes Risk decoding.