SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thought Models
Latent chain-of-thought models move intermediate reasoning from emitted text into continuous states, improving compactness but hiding the causal object. We introduce SCIT, the Suffix Cache Interchange Test, a causal protocol that constructs exact source-recipient counterfactuals, patches declared cache segments, and identifies which transformer object carries the counterfactual computation. SCIT combines sufficiency tests with K/V component splits, hidden-state controls, semantic source controls, decoded validation, and matched corruption. On CODI-GPT2 and a Sim-CoT-style GPT-2 reproduction, c
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 46%Chain of Thought is a scaling trap. the next wave is latent reasoning (Coconut / HRM / RecrusiveMAS)... but then we hit the black box wall. Where does BDH fit? [D] →
- FuzzySimilar title/name (fuzzy) · 87%LMCache/LMCache →
“Fuzzy title match (0.94): “SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thoug” ≈ “LMCache/LMCache””
- LinkedLinked via arxiv author · 85%Yi Ding →
“SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thought Models”
- LinkedLinked via arxiv author · 85%Lijun Huang →
“SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thought Models”
- LinkedLinked via arxiv author · 85%Menglin Yang →
“SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thought Models”