Optimizing Sparse Outcomes Through Dense Behavioral Signals via Value-Guided Preference Distillation

• URL: http://arxiv.org/abs/2609.14648v1
• Date: Sun, 13 Sep 2026 16:32:54 GMT
• Title: Optimizing Sparse Outcomes Through Dense Behavioral Signals via Value-Guided Preference Distillation
• Abstract summary: We formulate long-horizon dialogue optimization as a multi-objective reinforcement learning problem.<n>We show that a scalarized composite of dense auxiliary behavioral signals enables effective credit assignment and optimization of sparse outcomes.<n>We demonstrate that our distilled policy significantly improves long-term user retention, while simultaneously enhancing the positive behaviors and therapeutic-process markers.
• Score: 5.277024349608834
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Aligning multi-turn dialogue agents is usually framed as matching turn-level human preferences, yet direct optimization of long-term outcomes is often ineffective and prone to reward hacking. We formulate long-horizon dialogue optimization as a multi-objective reinforcement learning problem and train a multi-head value model that predicts a vector of observed user behaviors across multiple look-ahead horizons. Our findings demonstrate that a scalarized composite of dense auxiliary behavioral signals enables effective credit assignment and optimization of sparse outcomes. However, optimizing unconstrained single-objective proxies might induce policy degradations that are harmful when the agent is exposed to real users. To identify these failure modes prior to deployment, we establish a safety framework combining counterfactual user simulation with a validated dialogue-level outcome model to evaluate preference weightings and policy optimization methods. Finally, we demonstrate that distilling multi-objective value preferences into the policy via reference-anchored preference optimization matches on-policy online RL at a small fraction of its compute budget. Live A/B testing confirms that our distilled policy significantly improves long-term user retention, while simultaneously enhancing the positive behaviors and therapeutic-process markers.

Related papers

• Think2Go: Generative Next POI Recommendation with LLM Reasoning [57.37706163990998]
  Next Point-of-Interest (POI) recommendation task focuses on mining user behavioral preference patterns from historical check-ins to provide personalized suggestions for the next destination.<n>Existing methods rely on shallow contextual information and handcrafted feature interactions to predict the next POI.<n>We propose Think2Go, a novel generative next POI recommendation framework, which enhances the model's comprehension of SID representations and explores diverse spatial-temporal patterns via test-time computational scaling.
  arXiv  Detail & Related papers  (2026-07-31T03:52:20Z)
• Long-Term Optimization for Large-Scale Generative Retrieval with Off-Policy REINFORCE [0.0]
  We formulate recommendation as a session-level sequential decision-making problem.<n>We introduce an autoregressive approach for training generative retrievers with off-policy REINFORCE on pre-collected data.
  arXiv  Detail & Related papers  (2026-07-02T23:13:36Z)
• Trait-Aware Policy Optimization for Autoregressive Multi-Trait Essay Scoring [27.596544379317436]
  Trait-Aware Policy Optimization (TAPO) is a post-training framework tailored to autoregressive multi-trait scoring.<n>Our method decomposes rewards along both the sample and trait dimensions, combining global scoring consistency, trait-level accuracy, format validity, and inter-trait dependency preservation.
  arXiv  Detail & Related papers  (2026-05-25T11:41:37Z)
• Meta-Aligner: Bidirectional Preference-Policy Optimization for Multi-Objective LLMs Alignment [35.36029664224222]
  Multi-Objective Alignment aims to align Large Language Models with diverse and often conflicting human values.<n>We propose Meal, i.e., MEta ALigner, a bi-level meta-learning framework enabling bidirectional optimization between preferences and policy responses.
  arXiv  Detail & Related papers  (2026-04-27T08:36:13Z)
• GroupDPO: Memory efficient Group-wise Direct Preference Optimization [29.469340103121144]
  We introduce a memory-efficient group-wise preference optimization algorithm that preserves gradients while decoupling samples during backpropagation.<n>Across both offline and online alignment settings, we show that leveraging multiple responses consistently outperforms single-pair training.
  arXiv  Detail & Related papers  (2026-04-17T00:56:59Z)
• How Sampling Shapes LLM Alignment: From One-Shot Optima to Iterative Dynamics [65.67654005892469]
  We show that proper instance-dependent sampling can yield stronger ranking guarantees, while skewed on-policy sampling can induce excessive concentration under structured preferences.<n>We then analyze iterative alignment dynamics in which the learned policy feeds back into future sampling and reference policies.<n>Our theoretical insights extend to Direct Preference Optimization, indicating the phenomena we captured are common to a broader class of preference-alignment methods.
  arXiv  Detail & Related papers  (2026-02-12T17:11:08Z)
• RecNet: Self-Evolving Preference Propagation for Agentic Recommender Systems [109.9061591263748]
  RecNet is a self-evolving preference propagation framework for recommender systems.<n>It proactively propagates real-time preference updates across related users and items.<n>In the backward phase, the feedback-driven propagation optimization mechanism simulates a multi-agent reinforcement learning framework.
  arXiv  Detail & Related papers  (2026-01-29T12:14:31Z)
• On Symmetric Losses for Robust Policy Optimization with Noisy Preferences [55.8615920580824]
  This work focuses on reward modeling, a core component in reinforcement learning from human feedback.<n>We propose a principled framework for robust policy optimization under noisy preferences.<n>We prove that symmetric losses enable successful policy optimization even under noisy labels.
  arXiv  Detail & Related papers  (2025-05-30T15:30:43Z)
• On the Role of Feedback in Test-Time Scaling of Agentic AI Workflows [71.92083784393418]
  Agentic AI (systems that autonomously plan and act) are becoming widespread, yet their task success rate on complex tasks remains low.<n>Inference-time alignment relies on three components: sampling, evaluation, and feedback.<n>We introduce Iterative Agent Decoding (IAD), a procedure that repeatedly inserts feedback extracted from different forms of critiques.
  arXiv  Detail & Related papers  (2025-04-02T17:40:47Z)
• Understanding Likelihood Over-optimisation in Direct Alignment Algorithms [20.043560907227018]
  Direct Alignment Algorithms (DAAs) have emerged as alternatives to online Reinforcement Learning from Human Feedback. These algorithms aim to increase the likelihood of generating better (preferred) completions while discouraging worse (non-preferred) ones. This work explores the relationship between completion likelihood and model performance in state-of-the-art DAAs.
  arXiv  Detail & Related papers  (2024-10-15T15:14:22Z)
• An Efficient Continuous Control Perspective for Reinforcement-Learning-based Sequential Recommendation [14.506332665769746]
  We propose an underlinetextbfEfficient underlinetextbfContinuous underlinetextbfControl framework (ECoC) Based on a statistically tested assumption, we first propose the novel unified action representation abstracted from normalized user and item spaces. During this process, strategic exploration and directional control in terms of unified actions are carefully designed and crucial to final recommendation decisions.
  arXiv  Detail & Related papers  (2024-08-15T09:26:26Z)
• Bridging and Modeling Correlations in Pairwise Data for Direct Preference Optimization [75.1240295759264]
  We propose an effective framework for Bridging and Modeling Correlations in pairwise data, named BMC.<n>We increase the consistency and informativeness of the pairwise preference signals through targeted modifications.<n>We identify that DPO alone is insufficient to model these correlations and capture nuanced variations.
  arXiv  Detail & Related papers  (2024-08-14T11:29:47Z)
• Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizer [52.09480867526656]
  We identify the source of misalignment as a form of distributional shift and uncertainty in learning human preferences.<n>To mitigate overoptimization, we first propose a theoretical algorithm that chooses the best policy for an adversarially chosen reward model.<n>Using the equivalence between reward models and the corresponding optimal policy, the algorithm features a simple objective that combines a preference optimization loss and a supervised learning loss.
  arXiv  Detail & Related papers  (2024-05-26T05:38:50Z)

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.