Skip to content

EvoSkillRec: Skill-Genome Evolution for Recommender Architecture Discovery

Sep 2026 · arXiv.org · Vol abs/2609.34552 · 1 citation · 59 references
Computer Science

TL;DR

This work introduces EvoSkillRec, a promotion-and-reuse framework for cumulative recommender architecture evolution that decomposes recommenders into atomic executable skills and represents architectures as typed skill genomes, with each skill equipped with input--output types, semantic annotations, and implementation code.

Abstract

Modern recommender systems advance not only by scaling data and parameters, but also by encoding task-specific inductive biases through architecture, including sparse feature interactions for click-through rate (CTR) prediction, temporal attention for sequential recommendation, and expert routing for multi-task learning. However, these biases are typically human expert designed or searched within predefined operator spaces. Although Recent LLM-driven code evolution expands this space, unconstrained edits often produce invalid or ineffective architectures, underuse established architecture design knowledge, and fail to preserve successful innovations for reuse. We introduce EvoSkillRec, a promotion-and-reuse framework for cumulative recommender architecture evolution. It first decomposes recommenders into atomic executable skills and represents architectures as typed skill genomes, with each skill equipped with input--output types, semantic annotations, and implementation code. We then evolve models with different tasks through two coupled spaces: a constrained skill--space that mutates, recombines, specializes, and reuses validated skills, and an open-ended code--space in which LLM planners and synthesizers invent new skill modules using prior evolution traces and accumulated experience. An autoresearch controller evaluates candidates, diagnoses failures, retrieves relevant skills, promotes validated innovations into the skill library, and adaptively allocates the proposal budget between the two spaces. Extensive experiments on CTR prediction, multi-task learning, and multi-domain learning, including resource-constrained co-optimization of predictive quality and model FLOPs utilization in generative ranking models, consistently demonstrate the effectiveness of our proposed EvoSkillRec.

View source

Similar papers

Review Nov 2026

DualEvolve: Dual-Level Self-Evolution of Mechanism and Orchestration for Recommendation

Recommender systems are widely used in e-commerce and content platforms, where improvement requires repeated model design, training optimization, and experimental validation. Large language model (LLM) agents increasingly automate this development cycle. Sustained autonomous improvement requires agents to improve both...

Tian-Qi Gao, Chen-Xu Wang, Wen-Xuan Yang et al. · 0 citations
Sep 2026

Inherit4Rec: Parameter Inheritance for Efficient Scaling of Recommendation Models

The effectiveness of Inherit4Rec for continual capacity expansion and computation-efficient sparse conversion in industrial recommender systems is demonstrated, and both transformations consistently outperform the evaluated inheritance baselines across all prediction objectives.

Rui Zhang, Bo Chen, Xiao Wang et al. · 0 citations
#artificial intelligence Preprint Oct 2026

TGR-EvoDAG: A DAG-based Multi-Agent Code Evolution Framework for Generative Recommendation on Tencent

TGR-EvoDAG is proposed, a multi-agent code evolution framework tailored for GR that enables the first autonomous evolution of a complete GR pipeline, and introduces stage-aware cost planning to optimize GPU-hour budgets, and active downstream co-adaptation to navigate temporary regressions and unlock cross-stage gains.

Jun-Lin He, Jun-Zhang Zhu, Yu-Dong Li et al. · 0 citations
Book Open access Nov 2026

Path-based Reasoning on Knowledge Graphs for Explainable Recommender Systems

Recommender systems are central in modern knowledge-intensive applications. Beyond predicting user-item interactions, this class of systems increasingly requires models to reason over heterogeneous entities, relations, and contextual evidence while remaining transparent, trustworthy, and reusable. Path reasoning over k...

L. Boratto, Gianni Fenu, Francesca Maridina Malloci et al. · 0 citations
Book Open access Sep 2026

UniTraj: Cross-Domain Long-Sequence Modeling for Commercial Recommendation

Long-sequence modeling is increasingly important in recommender systems for capturing users’ evolving and long-term interests. In advertising, however, user interaction histories are often highly sparse due to limited exposure opportunities, making ad-only behavior sequences insufficient for effective long-sequence rec...

Xian Hu, Ming Yue, Zhi-Xiang Feng et al. · 0 citations
Preprint Oct 2026

RoFiGR: Coarse Routing, Fine Refinement via Level-Adaptive Distillation in Multimodal Generative Recommendation

Generative recommenders predict an item's semantic identifier (SID) one code at a time. Early codes select broad branches, so a routing error can exclude relevant items before fine-grained prediction begins. In multimodal recommendation, these coarse decisions must reflect collaborative preferences as well as item sema...

Ji-Nan Dai, Zhi-Peng Wei, Zhuang Zhuang et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.