specific method · filed under model architecture
Index Branch
A lightweight branch that scores key-value blocks and selects a Top-K subset for attention.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- core
How sources treat it
One count per evidence span, weakest treatment to strongest.
core 2
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
a lightweight Index Branch scores key–value blocks and independently selects a Top- subset for each GQA group
coremodel architecturein MiniMax-M3MiniMax
the Index Branch selects blocks of keys/values
coremodel architecturein MiniMax Sparse AttentionMiniMax
Filed alongside
Other methods under model architecture :: token mixer :: sparse attention :: sparse attention indexer.
Compressed Sparse Attention 2Hierarchical Sparse IndexerIndexCacheIndexShareLightning IndexerReindex ModeReuse ModeDense Warm-up StageFine-grained token selectionIndexer WarmupReuse QSA index selection across speculative decoding stepsAverage poolingBlock-causal scoringCompressed lightweight indexerCross-stage shared-state management for attention reuseDetached indexer-input optimizationIndex Branch outputIndex Branch value headIndexPoolMQA indexerSingle-head index key