Robust Post-Training for Generative Recommenders: Why Exponential Reward-Weighted SFT Outperforms RLHF

• URL: http://arxiv.org/abs/2603.10279v1
• Date: Tue, 10 Mar 2026 23:48:25 GMT
• Title: Robust Post-Training for Generative Recommenders: Why Exponential Reward-Weighted SFT Outperforms RLHF
• Abstract summary: Existing post-training methods reward hack due to noisy user feedback and unreliable reward models.<n>We identify exponential reward-weighted SFT with weights $w = exp(r/)$ as uniquely suited to this setting.<n>We prove the first policy improvement guarantees for this setting under noisy rewards.
• Score: 7.2858507889096815
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Aligning generative recommender systems to user preferences via post-training is critical for closing the gap between next-item prediction and actual recommendation quality. Existing post-training methods are ill-suited for production-scale systems: RLHF methods reward hack due to noisy user feedback and unreliable reward models, offline RL alternatives require propensity scores that are unavailable, and online interaction is infeasible. We identify exponential reward-weighted SFT with weights $w = \exp(r/λ)$ as uniquely suited to this setting, and provide the theoretical and empirical foundations that explain why. By optimizing directly on observed rewards without querying a learned reward model, the method is immune to reward hacking, requires no propensity scores, and is fully offline. We prove the first policy improvement guarantees for this setting under noisy rewards, showing that the gap scales only logarithmically with catalog size and remains informative even for large item catalogs. Crucially, we show that temperature $λ$ explicitly and quantifiably controls the robustness-improvement tradeoff, providing practitioners with a single interpretable regularization hyperparameter with theoretical grounding. Experiments on three open-source and one proprietary dataset against four baselines confirm that exponential reward weighting is simple, scalable, and consistently outperforms RLHF-based alternatives.

Related papers

• Exponential Reward Weighting for Fine-Tuning Generative Recommenders under Sparse and Noisy Feedback [7.2858507889096815]
  In recommendation systems, users interact with only a small fraction of a vast item catalog, producing feedback that is both sparse and noisy.<n>We propose Exponential reward-weighted fine-tuning (Exp-RSFT)<n>We show that Exp-RSFT's suboptimality decomposes into two costs: a coverage cost arising from limitations of the logging policy and a noise cost from imperfect feedback.
  arXiv  Detail & Related papers  (2026-08-01T18:48:39Z)
• PriFT: Prior-Support Guided Supervised Fine-Tuning [74.65198014829393]
  Supervised fine-tuning (SFT) is an efficient approach for downstream task adaptation.<n>SFT fits fixed demonstrations token by token, including targets poorly aligned with the model's pretrained distribution.<n>We propose PriFT, which derives token weights from a frozen pretrained reference to obtain a stable reweighting signal unaffected by fine-tuning.
  arXiv  Detail & Related papers  (2026-06-08T12:14:06Z)
• CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning [57.24524263804788]
  Code verifiers play a critical role in post-verification for LLM-based code generation.<n>Existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency.<n>We show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples.
  arXiv  Detail & Related papers  (2026-01-30T10:33:29Z)
• Information-Theoretic Reward Modeling for Stable RLHF: Detecting and Mitigating Reward Hacking [78.69179041551014]
  We propose an information-theoretic reward modeling framework based on the Information Bottleneck principle.<n>We show that InfoRM filters out preference-irrelevant information to alleviate reward misgeneralization.<n>We also introduce IBL, a distribution-level regularization that penalizes such deviations, effectively expanding the optimization landscape.
  arXiv  Detail & Related papers  (2025-10-15T15:51:59Z)
• Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers [90.50039419576807]
  Reinforcement Learning with Verifiable Rewards (RLVR) trains policies against automated verifiers to avoid costly human labeling.<n>To reduce vulnerability to verifier hacking, many RLVR systems collapse rewards to binary $0,1$ during training.<n>This choice carries a cost: it introduces textitfalse negatives (rejecting correct answers, FNs) and textitfalse positives (accepting incorrect ones, FPs)
  arXiv  Detail & Related papers  (2025-10-01T13:56:44Z)
• Towards Efficient Online Exploration for Reinforcement Learning with Human Feedback [12.158181906895186]
  Reinforcement learning with human feedback has emerged as a central paradigm for aligning large language models with human preferences.<n>We investigate exploration principles for online RLHF, where one seeks to refine both the reward model and the policy in a data-efficient manner.<n>Motivated by this insight, we propose a new exploration scheme that directs preference queries toward reducing uncertainty in reward differences.
  arXiv  Detail & Related papers  (2025-09-26T17:57:17Z)
• REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization [15.329281344012117]
  REINFORCE++ is a critic-free framework centered on textbfGlobal Advantage Normalization.<n>We introduce two variants: REINFORCE++, a highly efficient and general algorithm for general-domain RLHF, and REINFORCE++ /w baseline, a robust group-sampling variant for complex reasoning tasks.
  arXiv  Detail & Related papers  (2025-01-04T02:08:06Z)
• In-Dataset Trajectory Return Regularization for Offline Preference-based Reinforcement Learning [15.369324784520538]
  We propose In-Dataset Trajectory Return Regularization (DTR) for offline preference-based reinforcement learning.<n>DTR mitigates the risk of learning inaccurate trajectory stitching under reward bias.<n>We also introduce an ensemble normalization technique that effectively integrates multiple reward models.
  arXiv  Detail & Related papers  (2024-12-12T09:35:47Z)
• Value-Incentivized Preference Optimization: A Unified Approach to Online and Offline RLHF [80.32171988565999]
  We introduce a unified approach to online and offline RLHF -- value-incentivized preference optimization (VPO)<n>VPO regularizes the maximum-likelihood estimate of the reward function with the corresponding value function.<n>Experiments on text summarization and dialog verify the practicality and effectiveness of VPO.
  arXiv  Detail & Related papers  (2024-05-29T17:51:42Z)
• Robust Preference Optimization through Reward Model Distillation [68.65844394615702]
  Direct Preference Optimization (DPO) is a popular offline alignment method that trains a policy directly on preference data.<n>We analyze this phenomenon and use distillation to get a better proxy for the true preference distribution over generation pairs.<n>Our results show that distilling from such a family of reward models leads to improved robustness to distribution shift in preference annotations.
  arXiv  Detail & Related papers  (2024-05-29T17:39:48Z)
• SuperHF: Supervised Iterative Learning from Human Feedback [20.22920163075946]
  We focus on two prevalent methods used to align large language models, Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) We propose a novel approach, Supervised Iterative Learning from Human Feedback (SuperHF), which seeks to leverage the strengths of both methods. Our experimental results show SuperHF exceeds PPO-based RLHF on the training objective, easily and favorably trades off high reward with low reward hacking, improves downstream calibration, and performs the same on our GPT-4 based qualitative evaluation scheme all the while being significantly simpler to implement.
  arXiv  Detail & Related papers  (2023-10-25T16:52:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.