DICE-MMM is a two-stage diagnostic framework that separates forecasting accuracy from graph-aligned attribution in neural MMMs and localizes decoder bypass via controlled graph tests.
multi-hop leakage
5 Pith papers cite this work, alongside 253 external citations. Polarity classification is still indexing.
5
Pith papers citing it
253
external citations · external index
citation-role summary
background 2
citation-polarity summary
fields
cs.LG 5years
2026 5roles
background 2polarities
background 2representative citing papers
SVAR-FM uses simulator clamping to produce interventional distributions and flow matching to identify time series causal structures, with an error bound that predicts sign reversal of causal effects below a simulator accuracy threshold.
Mask2Cause recovers causal graphs directly during time-series forecasting via adjacency-constrained masked attention and achieves state-of-the-art discovery performance with over 70% reduction in forecasting parameters on average.
Integrates partial ODE physics into SDE-based causal discovery via drift-diffusion separation, with sparsity-inducing quasi-likelihood estimation, recovery guarantees for stable/unstable systems, and robustness analysis to model misspecification.
Anchoring personalized rolling-window causal graphs to a knowledge-guided population prior improves recovery of known dynamic health mechanisms and intervention directions on a semi-synthetic benchmark.
citing papers explorer
-
Forecasting Is Not Attribution: Localizing Decoder Bypass in Graph-Based Neural Marketing Mix Models
DICE-MMM is a two-stage diagnostic framework that separates forecasting accuracy from graph-aligned attribution in neural MMMs and localizes decoder bypass via controlled graph tests.
-
Intervention-Based Time Series Causal Discovery via Simulator-Generated Interventional Distributions
SVAR-FM uses simulator clamping to produce interventional distributions and flow matching to identify time series causal structures, with an error bound that predicts sign reversal of causal effects below a simulator accuracy threshold.
-
Mask2Cause: Causal Discovery via Adjacency Constrained Causal Attention
Mask2Cause recovers causal graphs directly during time-series forecasting via adjacency-constrained masked attention and achieves state-of-the-art discovery performance with over 70% reduction in forecasting parameters on average.
-
Causal Discovery from Heteroscedastic Stochastic Dynamical Systems under Imperfect Physical Models
Integrates partial ODE physics into SDE-based causal discovery via drift-diffusion separation, with sparsity-inducing quasi-likelihood estimation, recovery guarantees for stable/unstable systems, and robustness analysis to model misspecification.
-
PerCaM-Health: Personalized Dynamic Causal Graphs for Healthcare Reasoning
Anchoring personalized rolling-window causal graphs to a knowledge-guided population prior improves recovery of known dynamic health mechanisms and intervention directions on a semi-synthetic benchmark.