# ReCast: Recasting Learning Signals for Reinforcement Learning in Generative Recommendation

arXiv: https://arxiv.org/abs/2604.22169 · HTML: https://mlsanity.com/paper/2604.22169/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL with diffusion and flow models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 25 of its 31 references that are in the map.
- First posted on arXiv 24 Apr 2026.

## Placed in

- Deep reinforcement learning methods / [RL with diffusion and flow models](https://mlsanity.com/deeprl/rl-with-diffusion-and-flow-models/)

## Builds on (most cited first, 25 of 25)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)](https://mlsanity.com/paper/2203.13366/) (2022)
- [TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation](https://mlsanity.com/paper/2305.00447/) (2023)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation](https://mlsanity.com/paper/2311.09049/) (2023)
- [M6-Rec: Generative Pretrained Language Models are Open-Ended Recommender Systems](https://mlsanity.com/paper/2205.08084/) (2022)
- [Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems](https://mlsanity.com/paper/2504.01990/) (2025)
- [Learnable Item Tokenization for Generative Recommendation](https://mlsanity.com/paper/2405.07314/) (2024)
- [EAGER: Two-Stream Generative Recommender with Behavior-Semantic Collaboration](https://mlsanity.com/paper/2406.14017/) (2024)
- [OneRec-Think: In-Text Reasoning for Generative Recommendation](https://mlsanity.com/paper/2510.11639/) (2025)
- [Reinforced Latent Reasoning for LLM-based Recommendation](https://mlsanity.com/paper/2505.19092/) (2025)
- [Rec-R1: Bridging Generative Large Language Models and User-Centric Recommendation Systems via Reinforcement Learning](https://mlsanity.com/paper/2503.24289/) (2025)
- [OneRec-V2 Technical Report](https://mlsanity.com/paper/2508.20900/) (2025)
- [OneRec Technical Report](https://mlsanity.com/paper/2506.13695/) (2025)
- [Content-Based Collaborative Generation for Recommender Systems](https://mlsanity.com/paper/2403.18480/) (2024)
- [OpenOneRec Technical Report](https://mlsanity.com/paper/2512.24762/) (2025)
- [Reasoning over Semantic IDs Enhances Generative Recommendation](https://mlsanity.com/paper/2603.23183/) (2026)
- [OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation](https://mlsanity.com/paper/2512.22386/) (2025)
- [UNGER: Generative Recommendation with A Unified Code via Semantic and Collaborative Integration](https://mlsanity.com/paper/2502.06269/) (2025)
- [EAGER-LLM: Enhancing Large Language Models as Recommenders through Exogenous Behavior-Semantic Integration](https://mlsanity.com/paper/2502.14735/) (2025)
- [GFlowGR: Fine-tuning Generative Recommendation Frameworks with Generative Flow Networks](https://mlsanity.com/paper/2506.16114/) (2025)
- [Fine-grained Semantics Integration for Large Language Model-based Recommendation](https://mlsanity.com/paper/2602.22632/) (2026)
- [Spend Search Where It Pays: Value-Guided Structured Sampling and Optimization for Generative Recommendation](https://mlsanity.com/paper/2602.10699/) (2026)
- [SAGE: Sequence-level Adaptive Gradient Evolution for Generative Recommendation](https://mlsanity.com/paper/2601.21452/) (2026)
- [Towards Sample-Efficient and Stable Reinforcement Learning for LLM-based Recommendation](https://mlsanity.com/paper/2602.00632/) (2026)

## Abstract (arXiv)

Generic group-based RL assumes that sampled rollout groups are already usable learning signals. We show that this assumption breaks down in sparse-hit generative recommendation, where many sampled groups never become learnable at all. We propose ReCast, a repair-then-contrast learning-signal framework that first restores minimal learnability for all-zero groups and then replaces full-group reward normalization with a boundary-focused contrastive update on the strongest positive and the hardest negative. ReCast leaves the outer RL framework unchanged, modifies only within-group signal construction, and partially decouples rollout search width from actor-side update width. Across multiple generative recommendation tasks, ReCast consistently outperforms OpenOneRec-RL, achieving up to 36.6% relative improvement in Pass@1. Its matched-budget advantage is substantially larger: ReCast reaches the baseline's target performance with only 4.1% of the rollout budget, and this advantage widens with model scale. The same design also yields direct system-level gains, reducing actor-side update time by 16.60x, lowering peak allocated memory by 16.5%, and improving actor MFU by 14.2%. Mechanism analysis shows that ReCast mitigates the persistent all-zero / single-hit regime, restores learnability when natural positives are scarce, and converts otherwise wasted rollout budget into more stable policy updates. These results suggest that, for generative recommendation, the decisive RL problem is not only how to assign rewards, but how to construct learnable optimization events from sparse, structured supervision.