technique

Structural Causal Model

techniqueactiveprovisionalstructural-causal-model-e5e29526·1 events·first seen 22d ago

Aliases: Structural Causal Model

Co-occurring entities

all-edge F1 GPT-5.2-high causal discovery CausaLab

More like this (12)

causal generative modeling Chain-of-Causation causal discovery Pearl's Causal Hierarchy Latent Consistency Models Narrative Explanations Causal Semantic Guidance Russell Circumplex Model From Observation to Intervention: A Causal Audit of Expert Importance in Mixture-of-Experts Models A Causal Model of Theory of Mind in Conflict for Artificial Intelligence Tri-System Theory Moral Foundations Theory

Recent events (1)

6arXiv · cs.CL·22d ago·source ↗

CausaLab: Scalable Benchmark for Interactive Causal Discovery by LLM Agents

CausaLab is a new evaluation environment that tests LLM agents on interactive causal discovery tasks, requiring them to recover both causal graphs and structural equations from synthetic laboratory episodes governed by randomly sampled structural causal models (SCMs). The benchmark separates predictive accuracy from genuine causal understanding, revealing a persistent gap: GPT-5.2-high achieves 92% task accuracy in a 6-node observational setting but only 0.471 all-edge F1 for mechanism recovery. Mixed observation-intervention strategies improve structural fidelity, while pure intervention strategies underperform on both metrics. Premature stopping is identified as a key agent weakness, partially mitigated by prompting models to verify hypothesis-data consistency.

Evaluation and Benchmarking AI Safety Research all-edge F1 GPT-5.2-high causal discovery +3 more