Search papers, labs, and topics across Lattice.
To automate literature lineage mapping and prevent chronological inversions in AI citation graphs, the authors introduce EvoTree, a staged framework that decouples taxonomy induction from temporal refinement. The system combines graph-aware hierarchical clustering for backbone extraction with monotonic-path temporal fine-tuning that correctly anchors transitional papers to internal nodes, followed by topology-preserving LLM concept labeling. Evaluated on a new 11-subfield AI benchmark, EvoTree sets state-of-the-art results across NMI, citation-direction accuracy, and concept purity while uniquely succeeding at marginal-paper detection.
Standard taxonomy induction systematically corrupts AI research lineages by forcing transitional breakthroughs into leaf nodes鈥攁 failure mode resolved by enforcing monotonic temporal constraints over hierarchical citation graphs.
Surveys remain the primary way researchers grasp the lineage of methods within an AI subfield, but they scale poorly against the current rate of publication. Existing taxonomy-induction methods are largely leaf-bound and time-agnostic; they tend to force transitional papers into mature leaves and can create topological inversions between ancestors and descendants. We propose EvoTree, a staged framework that decouples conceptual backbone learning from temporal refinement: a graph-aware encoder with distribution-based hierarchical clustering yields a stable taxonomy backbone; temporal fine-tuning then re-attaches marginal papers to internal nodes under monotonic-path constraints; a final LLM pass labels concepts without altering the topology. We release the first annotated benchmark for this task across 11 AI subfields. EvoTree attains the highest NMI and citation-direction accuracy among all baselines and the best concept purity on the annotated benchmark, and is the only method with non-trivial marginal-paper detection on the annotated set.
