# ScaleToT: Generalizing Structured LLM Reasoning for Billion-Scale Low-Activity User Modeling

arXiv: https://arxiv.org/abs/2606.24605 · HTML: https://mlsanity.com/paper/2606.24605/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 26 of its 38 references that are in the map.
- First posted on arXiv 23 Jun 2026.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 26 of 26)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations](https://mlsanity.com/paper/2402.17152/) (2024)
- [Cross-Domain Recommendation: Challenges, Progress, and Prospects](https://mlsanity.com/paper/2103.01696/) (2021)
- [Towards Open-World Recommendation with Knowledge Augmentation from Large Language Models](https://mlsanity.com/paper/2306.10933/) (2023)
- [A Survey on User Behavior Modeling in Recommender Systems](https://mlsanity.com/paper/2302.11087/) (2023)
- [Review-driven Personalized Preference Reasoning with Large Language Models for Recommendation](https://mlsanity.com/paper/2408.06276/) (2024)
- [Multi-Behavior Sequential Recommendation with Temporal Graph Transformer](https://mlsanity.com/paper/2206.02687/) (2022)
- [Billion-user Customer Lifetime Value Prediction: An Industrial-scale Solution from Kuaishou](https://mlsanity.com/paper/2208.13358/) (2022)
- [LLMs for User Interest Exploration in Large-scale Recommendation Systems](https://mlsanity.com/paper/2405.16363/) (2024)
- [LettinGo: Explore User Profile Generation for Recommendation System](https://mlsanity.com/paper/2506.18309/) (2025)
- [Real-time Indexing for Large-scale Recommendation by Streaming Vector Quantization Retriever](https://mlsanity.com/paper/2501.08695/) (2025)
- [Modeling User Retention through Generative Flow Networks](https://mlsanity.com/paper/2406.06043/) (2024)
- [Diverse Preference Augmentation with Multiple Domains for Cold-start Recommendations](https://mlsanity.com/paper/2204.00327/) (2022)
- [STAN: Stage-Adaptive Network for Multi-Task Recommendation by Learning User Lifecycle-Based Representation](https://mlsanity.com/paper/2306.12232/) (2023)
- [Latent User Intent Modeling for Sequential Recommenders](https://mlsanity.com/paper/2211.09832/) (2022)
- [ADSNet: Cross-Domain LTV Prediction with an Adaptive Siamese Network in Advertising](https://mlsanity.com/paper/2406.10517/) (2024)
- [Cold-start Problems in Recommendation Systems via Contextual-bandit Algorithms](https://mlsanity.com/paper/1405.7544/) (2014)
- [DELRec: Distilling Sequential Pattern to Enhance LLMs-based Sequential Recommendation](https://mlsanity.com/paper/2406.11156/) (2024)
- [Efficient and Deployable Knowledge Infusion for Open-World Recommendations via Large Language Models](https://mlsanity.com/paper/2408.10520/) (2024)
- [MARS: Modality-Aligned Retrieval for Sequence Augmented CTR Prediction](https://mlsanity.com/paper/2509.01184/) (2025)
- [ChoirRec: Semantic User Grouping via LLMs for Conversion Rate Prediction of Low-Activity Users](https://mlsanity.com/paper/2510.09393/) (2025)
- [User-Adaptive Meta-Learning for Cold-Start Medication Recommendation with Uncertainty Filtering](https://mlsanity.com/paper/2601.22820/) (2026)

## Abstract (arXiv)

Accurate user modeling often depends on rich interaction histories, which are unavailable for billions of low-activity users. Large Language Models (LLMs) can infer latent user states from static profiles, but this reasoning becomes unreliable when profiles are sparse, and applying an LLM to billions of users is prohibitively expensive. We present ScaleToT, which learns structured reasoning from a small LLM-processed subset and extends it to the broader low-activity user population. To improve reasoning reliability, ScaleToT constructs typed user-state chains with a bounded entropy-guided Tree-of-Thought (ToT) refinement procedure. To make this structured reasoning usable from sparse profiles, the teacher-curated chains are used to train a student model on static profiles through supervised fine-tuning (SFT) and Outcome-Driven Segment-Aware Implicit Reward Policy Optimization (OSIPO). ScaleToT then transfers the student's reasoning representations to a lightweight profile encoder, providing shared reasoning signals for the remaining users without LLM inference. We evaluate ScaleToT on lifetime value (LTV) prediction in a billion-scale advertising deployment. A randomized online A/B test increased LT30 by 6.738%, while offline reasoning covered only 7.32% of the potential population, greatly reducing compute cost compared with full-population reasoning.