Towards Automatic Evolution Tree Generation from Citation Graphs

2026-09-08

AI Summary

为解决现有方法无法有效处理大量论文的问题，提出EvoTree框架，通过图感知编码器和时序微调生成稳定的进化树，并在11个AI子领域进行了验证。

Abstract

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.

Problem

• taxonomy-induction
• temporal-agnostic
• topological inversions

Innovation

• EvoTree
• graph-aware encoder
• temporal refinement
• hierarchical clustering
• monotonic-path constraints