GRAFT: Graph-Matched Retrieval and Fusion of Tables in Data Lakes
For autonomous data agents and data lake systems, GRAFT provides a novel graph-matching approach that significantly improves table retrieval and fusion accuracy.
GRAFT addresses the problem of retrieving and integrating tables from data lakes for analytical queries, where existing methods ignore joinability and unionability. It achieves best Recall, Precision, F1, and Sufficiency on Spider and BIRD benchmarks, with relative gains of 7.8% in F1 and 10.6% in Sufficiency over the strongest baseline.
Autonomous data agents resolve analytical queries by retrieving and reasoning over evidence in tabular data lakes. Existing methods score tables independently against the query and ignore the joinability and unionability that link them, returning fragmented evidence that downstream agents cannot integrate. We propose GRAFT (Graph-matched Retrieval and Fusion of Tables), structured around two principal contributions. First, we cast table retrieval as a graph matching problem between a query-derived intent graph and a heterogeneous data lake graph, and introduce IGMS, a log-determinant reward that couples semantic relevance, structural compatibility, and evidence diversity in a single objective. Second, we recast subgraph generation as a Markov decision process and learn a value function via implicit Q-learning on self-generated trajectories produced by a canonical compression operator that inverts the homomorphism. We further design a three-stage online pipeline that exploits anchor reachability, predicate admissibility, and reward monotonicity to greatly prune the candidate space before exact IGMS evaluation. On Spider and BIRD adapted to the tabular data lake setting, GRAFT achieves the best Recall, Precision, F1, and Sufficiency among point-wise, greedy-expansion, and structure-aware baselines, with relative gains of 7.8% in F1 and 10.6% in Sufficiency over the strongest baseline, while maintaining high search efficiency.