specific method · filed under inference & serving
Summary-based context compression
A context-management strategy identified as summary-based compression, distinct from resetting the context with discard-all.
Also called summary-based compression.
- source
- 1
- model
- 1
- labs adopt it
- 0
- strongest
- evaluated
How sources treat it
One count per evidence span, weakest treatment to strongest.
evaluated 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
including discard-all resets and summary-based compression.
evaluatedunclearin Search teacher for Nemotron 3 UltraNVIDIA
Filed alongside
Other methods under inference & serving :: context management.
Discard-all context managementPreserved thinking history modeExcluding prior thinking from conversation historyContext compactionContext foldingDiscard-75%Preserve thinkingThinking context management for tool useTrajectory summarization and rollout re-initiationContext management methodContext management strategyDiscarding tool-call historyHierarchical context managementKeep-recent-kMemory compressionModality-specific deploymentTest-time context management for extending token budgets