A new causal modeling diagram adds ordering, context, adjacency, and mechanism variables
KordingLab · x · 2026-07-27
- The post says the authors are trying to broaden the modeling of causality.
- The accompanying diagram shows a two-phase causal graph with exogenous ordering, context variables, adjacency variables, mechanism variables, and endogenous variables.
- The figure suggests a gated causal-edge mechanism where adjacency controls which edges are active across phases.
More from Research
- A computer science paper titled “In Cantor Space No One Can Hear You Stream” — fkasummer · 2026-07-27
- YC talk says robots need better pretraining, memory, and compositional behaviors — ycombinator · 2026-07-27
- Hugging Face dataset turns Bay Bridge traffic into an hourly historical explorer — jwt0625 · 2026-07-27
- Karpathy-style graph engineering turns multi-agent loops into shared-memory systems — leslysandra · 2026-07-27
- Google’s JAXBench benchmark lifts TPU kernel optimization with 50 real workloads — omarsar0 · 2026-07-27
- AgentPrune cuts multi-agent LLM communication costs from $43.7 to $5.6 in a new paper — sebkrier · 2026-07-27