# A Better Spur Should Start From Each Objective

arXiv: https://arxiv.org/abs/2609.08211 · HTML: https://mlsanity.com/paper/2609.08211/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods.
- Builds on 14 of its 22 references that are in the map.
- First posted on arXiv 8 Sep 2026.

## Placed in

- Deep reinforcement learning methods / [RL for games and combinatorial optimization](https://mlsanity.com/deeprl/rl-for-games-and-combinatorial-optimization/)

## Builds on (most cited first, 14 of 14)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains](https://mlsanity.com/paper/2006.10739/) (2020)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [ToolRL: Reward is All Tool Learning Needs](https://mlsanity.com/paper/2504.13958/) (2025)
- [GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization](https://mlsanity.com/paper/2601.05242/) (2026)
- [Imbalanced Gradients in RL Post-Training of Multi-Task LLMs](https://mlsanity.com/paper/2510.19178/) (2025)
- [M3HF: Multi-agent Reinforcement Learning from Multi-phase Human Feedback of Mixed Quality](https://mlsanity.com/paper/2503.02077/) (2025)
- [Tree-based Dialogue Reinforced Policy Optimization for Red-Teaming Attacks](https://mlsanity.com/paper/2510.02286/) (2025)
- [DARE-bench: Evaluating Modeling and Instruction Fidelity of LLMs in Data Science](https://mlsanity.com/paper/2602.24288/) (2026)
- [Making MLLMs Blind: Adversarial Smuggling Attacks in MLLM Content Moderation](https://mlsanity.com/paper/2604.06950/) (2026)
- [Beyond Scalar Rewards: An Axiomatic Framework for Lexicographic MDPs](https://mlsanity.com/paper/2505.12049/) (2025)
- [Meta-Aligner: Bidirectional Preference-Policy Optimization for Multi-Objective LLMs Alignment](https://mlsanity.com/paper/2604.24178/) (2026)

## Built on by (most cited first, 1 of 1)

- [How Can Recommendation Feedback Evolve Agent Memory?](https://mlsanity.com/paper/2609.37544/) (2026)

## Abstract (arXiv)

Real-world Multi-Objective Reinforcement Learning (MORL) often suffers from sparse rewards, reward conflicts, and late-stage reward tug-of-war, causing traditional linear scalarization to experience severe metric oscillations. To address optimization conflicts among multiple objectives in real-world deployment scenarios, we propose Multi-Marginal Preference Optimization (MMPO), a fine-grained framework that intervenes at the data, gradient, and constraint levels rather than relying on coarse-grained global scalarization. Specifically, MMPO performs exposure debiasing to mitigate sparse and biased rewards, applies priority-aware orthogonal projection to decouple conflicting gradients, and introduces self-prompted gradient constraints to prevent dominant objectives from overwhelming weaker ones. Experiments on real-world e-commerce datasets show that MMPO improves training stability and consistently achieves better performance across conflicting metrics. Moreover, it generalizes robustly to broader tasks such as ToolRL and code generation, demonstrating its effectiveness as a practical paradigm for multi-objective alignment.