Task Expansion and Cross Refinement for Open-World Conditional Modeling
This addresses the challenge of limited task coverage in open-world conditional modeling for applications requiring flexible querying across diverse datasets, representing an incremental advancement through a novel framework.
The paper tackles the problem of open-world conditional modeling, where a single model must handle arbitrary conditional queries across heterogeneous datasets with varying observed variables and targets, by proposing TEXR, a semi-supervised framework that expands task coverage through structured synthesis and refinement, resulting in improved zero-, few-, and many-shot performance across heterogeneous tabular benchmarks.
Open-world conditional modeling (OCM), requires a single model to answer arbitrary conditional queries across heterogeneous datasets, where observed variables and targets vary and arise from a vast open-ended task universe. Because any finite collection of real-world datasets covers only a small fraction of this space, we propose Task Expansion and Cross Refinement (TEXR), a semi-supervised framework that enlarges effective task coverage through structured synthesis and refinement of semantic data contexts. TEXR first generates diverse uninstantiated dataset schemas and weakly instantiates them via structured probabilistic generators guided by large language models. It then performs cross-model refinement by training on disjoint data partitions and revising synthetic values across splits to reduce confirmation bias and improve pseudo-value quality. The refined synthetic datasets are aggregated with real data to train a unified conditional model. Across heterogeneous tabular benchmarks, TEXR consistently improves zero-, few-, and many-shot performance for multiple OCM backbones, demonstrating that structured task expansion and cross refinement enhance open-world conditional modeling.