AgentLTV: An Agent-Based Unified Search-and-Evolution Framework for Automated Lifetime Value Prediction

Chaowei Wu; Huazhu Chen; Congde Yuan; Qirui Yang; Guoqing Song; Yue Gao; Li Luo; Frank Youhua Chen

cs.LGFeb 25, 2026

AgentLTV: An Agent-Based Unified Search-and-Evolution Framework for Automated Lifetime Value Prediction

Authors: Chaowei Wu, Huazhu Chen, Congde Yuan, Qirui Yang, Guoqing Song, Yue Gao, Li Luo, Frank Youhua Chen,

Organizations: Harbin Institute of Technology, Harbin, Heilongjiang, China · Xiangtan University, Xiangtan, Hunan, China

Abstract

Lifetime Value (LTV) prediction is critical in advertising, recommender systems, and e-commerce. In practice, LTV data patterns vary across decision scenarios. As a result, practitioners often build complex, scenario-specific pipelines and iterate over feature processing, objective design, and tuning. This process is expensive and hard to transfer. We propose AgentLTV, an agent-based unified search-and-evolution framework for automated LTV modeling. AgentLTV treats each candidate solution as an {executable pipeline program}. LLM-driven agents generate code, run and repair pipelines, and analyze execution feedback. Two decision agents coordinate a two-stage search. The Monte Carlo Tree Search (MCTS) stage explores a broad space of modeling choices under a fixed budget, guided by the Polynomial Upper Confidence bounds for Trees criterion and a Pareto-aware multi-metric value function. The Evolutionary Algorithm (EA) stage refines the best MCTS program via island-based evolution with crossover, mutation, and migration. Experiments on a large-scale proprietary dataset and a public benchmark show that AgentLTV consistently discovers strong models across ranking and error metrics. Online bucket-level analysis further indicates improved ranking consistency and value calibration, especially for high-value and negative-LTV segments. We summarize practitioner-oriented takeaways: use MCTS for rapid adaptation to new data patterns, use EA for stable refinement, and validate deployment readiness with bucket-level ranking and calibration diagnostics. The proposed AgentLTV has been successfully deployed online.

Figures & tables

Figure 1. The current challenges faced by the LTV prediction scene. The current challenges

Figure 2. AgentLTV overview. (a) Configuration module instantiates LLM-driven agents and execution interfaces. (b) MCTS module performs broad exploration over LTV pipeline programs via PUCT-guided expansion after the Code Generator generates the initial node algorithm. (c) EA module refines the best MCTS program using island-based evolution with crossover, mutation, and migration based on the previous optimal algorithm. Model Architecture

Period	Samples	Positive	Negative	Average LTV
Period 1	426,412	188,047	146,686	23.810
Period 2	382,081	171,554	132,201	25.323

Table 1. Basic statistics of each period dataset.

Figure 3. Monte Carlo tree search Module visualization. Visualization of MCTS module. Note that node indices in the first layer reflect the prior ordering of initial programs: smaller indices correspond to more mainstream deep learning architectures, whereas larger indices represent earlier machine learning baselines.

Model	Period 1				Period 2
Model	ER	Norm GINI	Spearman	RMSE	ER	Norm GINI	Spearman	RMSE
Wide&Deep	2.225 $\pm$ 0.097	0.623 $\pm$ 0.019	0.101 $\pm$ 0.036	264.92 $\pm$ 2.43	1.976 $\pm$ 0.059	0.648 $\pm$ 0.029	0.093 $\pm$ 0.029	274.01 $\pm$ 1.75
DeepFM	1.602 $\pm$ 0.098	0.409 $\pm$ 0.162	0.186 $\pm$ 0.072	264.48 $\pm$ 3.15	1.731 $\pm$ 0.222	0.384 $\pm$ 0.242	0.133 $\pm$ 0.104	291.07 $\pm$ 7.77
ZILN	1.492 $\pm$ 0.174	0.310 $\pm$ 0.450	0.323 $\pm$ 0.216	261.83 $\pm$ 4.07	1.719 $\pm$ 0.161	0.034 $\pm$ 0.311	0.189 $\pm$ 0.161	287.64 $\pm$ 1.69
DCN	2.625 $\pm$ 0.166	0.602 $\pm$ 0.025	0.030 $\pm$ 0.049	284.11 $\pm$ 9.87	2.392 $\pm$ 0.126	0.606 $\pm$ 0.037	0.028 $\pm$ 0.036	295.41 $\pm$ 9.84
GateNet	1.949 $\pm$ 0.116	0.661 $\pm$ 0.018	0.147 $\pm$ 0.020	257.24 $\pm$ 3.68	1.804 $\pm$ 0.044	0.689 $\pm$ 0.015	0.201 $\pm$ 0.013	270.69 $\pm$ 3.57

Table 2. Performance comparison with different approaches

Method	ER	Norm GINI	Spearman	RMSE
$w/o$ MCTS	1.030	0.778	0.633	273.42
$w/o$ EA	0.670	0.958	0.823	185.95
AgentLTV	0.594	0.962	0.817	141.09

Table 3. The ablation study of MCTS Module and EA Module.

Panel of Figure 4 — Figure 4. Online economic analysis on bucketed LTV values. (a) Same Interval Ratio: the fraction of users whose predicted LTV falls into the same value bucket as the bucket defined by the ground-truth LTV ranking. Higher is better and indicates stronger ranking consistency across buckets. (b) Mean Absolute Error (MAE) within each bucket: the absolute prediction error averaged over users in that bucket. Lower is better and reflects better value calibration.

Samples	Standard deviation	Average LTV
1,797,813	6,452.395	797.358

Table 4. Basic statistics of the public dataset.

Method	ER	Norm GINI	Spearman	RMSE
Wide&Deep	0.522	0.695	0.611	6306.097
DeepFM	0.533	0.697	0.615	4566.259
ZILN	0.504	0.726	0.645	4879.940
DCN	0.619	0.507	0.613	10833.894
GateNet	0.480	0.725	0.640	4938.126
Kuaishou	0.542	0.728	0.645	9138.638

Table 5. Results of comparative experiments on public datasets.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Rank	Gen	Node Value	Gini	Spearman	RMSE	ErrorRate	Evolution type	Method
1	gen_21	0.834602	0.9557	0.8140	97.71	0.7726	Crossover	enhanced_temporal_attention_mechanism
2	gen_4	0.832967	0.9540	0.8272	104.13	0.9014	Crossover	enhance_tcn_architecture_with_attention
3	gen_19	0.821321	0.9532	0.8168	111.36	0.6851	Crossover	hybrid_dual_branch_attention_tcn
4	gen_18	0.819654	0.9508	0.8292	117.27	0.7690	Crossover	kinematic_attention_mtl_rank_boost
5	gen_45	0.788169	0.9586	0.8148	125.33	0.6690	Mutation	log_cosh_robust_mtl_with_enhanced_features
6	gen_35	0.786940	0.9519	0.8148	124.57	0.9622	Crossover	ranking_loss_and_attention

Appendix

Table 6. EA Module top 20 LTV Model Results Ranked by Node Value

\mathcal{L}=\text{AWMSE}+0.35(\text{PinballLoss}+\text{BCE})+\lambda(1-\rho)

Appendix

Algorithm 3 The optimal model generated by AgentLTV

Automated Machine Learning

Automated machine learning (AutoML) builds and tunes data-to-model pipelines with minimal manual effort, balancing predictive quality, compute, and deployment constraints.

See all 96 papers

EvoRank: LLM-Guided Evolution of Multi-Objective Learning-to-Rank Pipelines
Aug 30, 2026Rayhan Patel, Shabaz PatelLearning to Rank Automated Algorithm Discovery
University of Maryland, College Park, MD, USA · Independent Researcher
Agentic ML Exploration (A-MLE) for Ads Ranking
Sep 8, 2026Erwin Gao, Vinodh Kumar Sunkara, Jingyi Guan +36Learning to Rank LLM Agent Evaluation
Meta Platforms, Inc. Menlo Park and New York, USA
TGR-EvoDAG: A DAG-based Multi-Agent Code Evolution Framework for Generative Recommendation on Tencent
Oct 23, 2026Junlin He, Junzhang Zhu, Yudong Li +5Generative Recommendation Large Language Model-Guided Optimization
Platform and Content Group, Tencent, Shenzhen, China · The Hong Kong Polytechnic University, Hong Kong, China · McGill University, Montreal, Canada +1

Aug 30, 2026J/K move · Enter open · S save

EvoRank: LLM-Guided Evolution of Multi-Objective Learning-to-Rank Pipelines

Rayhan Patel, Shabaz Patel

University of Maryland, College Park, MD, USA · Independent Researcher

Learning to Rank Automated Algorithm Discovery Large Language Model-Guided Optimization Multi-Objective Evolutionary Optimization

Open paper AskPDF arXiv

We present EvoRank, an open autonomous ranking engineer: an LLM-guided evolutionary loop that discovers complete Learning-to-Rank pipelines (features, models, losses, ensembles) for multi-objective e-commerce search. On the Expedia ICDM 2013 dataset, with relevance, conversion, and revenue as competing objectives, three independent runs each converge within 50 iterations (about ten dollars) on interpretable pipelines that beat an Optuna-tuned LambdaMART on 60k held-out queries, an advantage that persists at full data scale and places in the top 6 percent of the original competition. A first campaign, evolving only training objectives, builds the central design rule: it appeared to work on its selection fold (the small dataset it uses to pick winners) while a transfer audit, re-scoring winners on held-out data, showed the gains were almost entirely fitness noise (the randomness of its own scoring), and neither seeded domain knowledge nor richer diagnostic feedback changed what transferred. The deciding quantity is measurable in advance: search-space headroom relative to fitness noise. We package this as a headroom gate that predicts, before any LLM spend, whether the loop will pay off, and we release the system, the auditing tools, and a catalog of failure modes with their guardrails, so teams can apply the procedure to their own ranking stacks.

Abstract

Automated Machine Learning

Explore similar work

EvoRank: LLM-Guided Evolution of Multi-Objective Learning-to-Rank Pipelines

Agentic ML Exploration (A-MLE) for Ads Ranking

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