Unverified50% confidenceConceptExact time
Long-context capability can be decomposed into three categories: recall, state tracking and pattern learning.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSAGA distills interaction trajectories layer by layer into three levels of hierarchical knowledge: episodic descriptions, reusable procedures, and principles with explicit applicability conditions.74% similarUnverifiedStateful tool agents maintain persistent state across multi-turn interactions, with each tool call result written into the execution trajectory and influencing subsequent decisions, giving their errors a cascading effect.73% similarUnverified现代行为遥测系统通常包含三个层次：事件层、状态层和流层72% similarUnverifiedThe orchestration layer of a dynamic workflow is responsible for three things: task decomposition, agent routing and result aggregation.70% similarUnverifiedKey technologies for long-horizon agents include state representation learning, goal alignment mechanisms, and error detection and recovery strategies (checkpoint mechanisms, sub-goal verification, and rollback strategies)69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074001API
curl https://kongchang.com/api/v1/knowledge/claims/1074001MCP
get_claim(id=1074001)