mlsanityData to 7 Oct 2026

PROMISE: Process Reward Models Unlock Test-Time Scaling Laws in Generative Recommendations

arXiv 2601.04674 ↗First posted 8 Jan 2026cs.IRcited by 2

On the map: LLM reasoning → LLM reasoning for recommendation

Abstract (arXiv)

Generative Recommendation has emerged as a promising paradigm, reformulating recommendation as a sequence-to-sequence generation task over hierarchical Semantic IDs. However, existing methods suffer from a critical issue we term Semantic Drift, where errors in early, high-level tokens irreversibly divert the generation trajectory into irrelevant semantic subspaces. Inspired by Process Reward Models (PRMs) that enhance reasoning in Large Language Models, we propose Promise, a novel framework that integrates dense, step-by-step verification into generative models. Promise features a lightweight PRM to assess the quality of intermediate inference steps, coupled with a PRM-guided Beam Search strategy that leverages dense feedback to dynamically prune erroneous branches. Crucially, our approach unlocks Test-Time Scaling Laws for recommender systems: by increasing inference compute, smaller models can match or surpass larger models. Extensive offline experiments and online A/B tests on a large-scale platform demonstrate that Promise effectively mitigates Semantic Drift, significantly improving recommendation accuracy while enabling efficient deployment.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in LLM reasoning for recommendation, one of 17 directions in LLM reasoning. Builds on 20 of its 49 references that are in the map. First posted on arXiv 8 Jan 2026.

Builds on

20 of 49 references.

References · 20

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,928
  4. 2022Solving math word problems with process- and outcome-based feedbackcited by 871
  5. 2024DeepSeek-V3 Technical Reportcited by 724
  6. 2023Finite Scalar Quantization: VQ-VAE Made Simplecited by 425
  7. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 402
  8. 2024Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoningcited by 267
  9. 2025OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignmentcited by 167
  10. 2025Process Reward Models That Thinkcited by 118
  11. 2024Entropy-Regularized Process Reward Modelcited by 38
  12. 2025Rec-R1: Bridging Generative Large Language Models and User-Centric Recommendation Systems via Reinforcement Learningcited by 32
8 more
  1. 2025OneRec-V2 Technical Reportcited by 28
  2. 2025OneRec Technical Reportcited by 24
  3. 2025OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Searchcited by 24
  4. 2025LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendationcited by 9
  5. 2024KuaiFormer: Transformer-Based Retrieval at Kuaishoucited by 6
  6. 2025OneLoc: Geo-Aware Generative Recommender Systems for Local Life Servicecited by 5
  7. 2025GRank: Towards Target-Aware and Streamlined Industrial Retrieval with a Generate-Rank Frameworkcited by 4
  8. 2024CRM: Retrieval Model with Controllable Conditioncited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026The Disconnect Between Better Descriptive Reasoning Trace Quality and Recommendation Effectivenesscited by 0
  2. 2026FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Learning from Unreachable Rewards: Hint-Conditioned Reinforcement Learning for Generative Recommendation
  2. 2026S2GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation
  3. 2026LASAR: Latent Adaptive Semantic Aligned Reasoning for Generative Recommendation
  4. 2026TwiSTAR:Think Fast, Think Slow, Then Act,Generative Recommendation with Adaptive Reasoning
  5. 2026Implicit Reasoning for Large Language Model-based Generative Recommendation
  6. 2025REG4Rec: Reasoning-Enhanced Generative Model for Large-Scale Recommendation Systems
  7. 2026Learning Better Reasoning for Generative Recommendation with Semantic IDs
  8. 2026GEMs: Breaking the Long-Sequence Barrier in Generative Recommendation with a Multi-Stream Decoder