Skip to content
#edge computing Book Open access

Context-Aware Pruning for Ordering-Based Causal Discovery

Nov 2026 · Proceedings of the 35th ACM International Conference on Information and Knowledge Management · pp. 4419-4429 · 0 citations · 6 references
Bayesian Modeling and Causal Inference

Abstract

Ordering-based causal discovery decomposes DAG learning into topological order estimation and edge pruning. Existing work has focused on order estimation, while pruning often relies on component-wise additive tests that can miss parent effects expressed through co-parent context. We propose Context-Aware Pruning of Edges (CAPE), a plug-in pruning module that formulates post-ordering edge pruning as local model selection. For each candidate edge, CAPE computes the out-of-sample gain in conditional log-predictive density of the child when the candidate parent is included in the current parent context. It retains the edge only when this contextual gain exceeds an MDL-inspired adaptive structural penalty. This criterion evaluates conditional predictive evidence rather than isolated additive effects, enabling CAPE to retain interaction-driven dependencies while pruning weak finite-sample edges. Across six ordering backbones, CAPE improves SHD, SID, and F1 on synthetic nonlinear graphs and assumption-violation benchmarks, including simulated latent confounding, measurement error, post-nonlinear mechanisms, and non-i.i.d. sampling. On empirical benchmarks, CAPE lowers average SHD and SID under the same estimated orders. Its hierarchical group-pruning strategy reduces pruning time by up to 7.5 times on larger graphs. Source code is available at https://github.com/ku-milab/CAPE.

Read PDF

Similar papers

#machine learning Preprint Oct 2026

Directional Evidence Guided Search-Space Reduction for Exact DAG Learning

Learning a directed acyclic graph (DAG) from observational data is a challenging combinatorial problem due to the exponential growth in the number of candidate parent-set configurations. Existing exact score-based methods often require computationally intensive combinatorial search, whereas constraint-based methods can...

U. J. Islam, A. Elrefaey, Rong Pan · 0 citations
Preprint Sep 2026

Functional Causal Discovery via Conditional Covariance Ordering

We study causal discovery where each node is a random function. Previous studies on this topic rely on structural assumptions, e.g., linearity or non-linearity, and distributional assumptions, e.g., Gaussianity or non-Gaussianity. In contrast, we make use of covariance operators to avoid these assumptions. Under functi...

Ke-Yu Li, Ruo-Xu Tan · 0 citations

Active Causal Discovery Benchmark: Evaluating LLM Agents Under Budgeted Interventions

We introduce the Active Causal Discovery Benchmark (ACDB), an SCM-grounded environment for evaluating whether LLM agents recover causal graph structure from observations and budget-constrained hard interventions. ACDB pairs a linear-Gaussian world generator with a fixed observe-intervene-submit API and a three-layer sc...

Sagar Deb, Devam Shah, Ashwanth Krishnan · 3 citations
Preprint Oct 2026

Residual-Based Causal Discovery via Pretrained In-Context Quantile Regression

Residual-based causal discovery assumes that each variable is generated from its parents and an independent noise source. It estimates this noise and tests its independence from candidate causes. For identifiable functions such as additive-noise models (ANMs) and the broader class of location-scale noise models (LSNMs)...

Daniel Klippert, Quang Nhat Huynh, Alexander Marx · 0 citations
#artificial intelligence Preprint Oct 2026

Causal Chain of Thought: Mitigating Attention Leakage via Context Isolation

Chain-of-thought (CoT) prompting elicits step-by-step rationales that appear to explain model predictions, but there is no guarantee these steps causally drive the final answer. In standard autoregressive generation, dense attention allows each reasoning step to attend to the full context, enabling the model to derive...

Gabriele Dominici, T. McMillan, Fazl Barez et al. · 0 citations

Related blog posts

Microsoft Research Blog Oct 29, 2026

Microsoft Research Focus – Edition #97

Host Sowbhagya Suresh shares some of the latest news and cutting-edge research from across Microsoft—from understanding the challenges facing coding agents to uncovering how people use AI at work and improving extended-range weather forecasting. The post Microsoft Research Focus – Edition #97 appeared first on Microsoft Research.

MIT News · Artificial Intelligence Oct 28, 2026

Making machines that can listen

Professor Paris Smaragdis develops computers with superhuman hearing for applications like speech recognition, music understanding, and environmental monitoring.

Microsoft Research Blog Oct 21, 2026

Solving a 31-year-old grand challenge in mathematics – the KLS conjecture

For decades, a fundamental question about the difficulty of computing in high-dimensional spaces has resisted proof. New research resolves this question, removing a major theoretical obstacle to understanding the efficiency of randomized algorithms used across science and computing. The Kannan–Lovász–Simonovits (KLS) conjecture was proposed in 1995. It asks whether, for a broad class of high-dimensional […] The post Solving a 31-year-old grand challenge in mathematics – the KLS conjecture appear…

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.