Eagle and Finch enhance RWKV with matrix-valued states and dynamic recurrence, trained on a 1.12-trillion-token multilingual corpus, and report competitive performance on standard benchmarks.
Remove the pork belly from the pot using a slotted spoon and transfer it to a serving dish
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence
Eagle and Finch enhance RWKV with matrix-valued states and dynamic recurrence, trained on a 1.12-trillion-token multilingual corpus, and report competitive performance on standard benchmarks.