Unverified50% confidenceFactExact time
During generation, SMC maintains a set of weighted candidate sequences (particles), reweighting and resampling them at each step according to the potential functions.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedState machine replication (SMR) is a core abstraction in distributed systems: multiple replicas start from the same initial state and execute the same deterministic command sequence, thereby keeping their states consistent.67% similarUnverifiedThe design of SMC is inspired by Speculative Decoding technology and elevates it from the token level to the action-chain level.66% similarUnverifiedA potential function is a scoring function in SMC that defines "how close the current partial sequence is to the goal"; it need not be differentiable or trained, and only needs to be computable in code.64% similarUnverifiedSA (Speculative Action) only performs single-step speculation, whereas SMC achieves reuse of multi-step speculative execution through the Macro Library, thereby delivering additional gains beyond single-step speculation.62% similarUnverifiedThe Macro Library in SMC mines and stores repeatedly occurring multi-action skeletons from training trajectories, and at runtime matches and reuses them against action chains predicted by the drafter model.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1115488API
curl https://kongchang.com/api/v1/knowledge/claims/1115488MCP
get_claim(id=1115488)