Pareto-Optimal Offline Reinforcement Learning via Smooth Tchebysheff Scalarization

• URL: http://arxiv.org/abs/2604.13175v1
• Date: Tue, 14 Apr 2026 18:00:39 GMT
• Title: Pareto-Optimal Offline Reinforcement Learning via Smooth Tchebysheff Scalarization
• Abstract summary: Large language models can be aligned with human preferences through offline reinforcement learning (RL) on small labeled datasets.<n>We use this to derive Smooth Tchebysheff of Multi-Objective Preferences (STOMP), a novel formulation that extends direct optimization to the multiobjective.<n>STOMP achieves the highest hyper-objectives in eight of nine settings according to both off-policy and generative evaluations.
• Score: 2.5513673603819886
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models can be aligned with human preferences through offline reinforcement learning (RL) on small labeled datasets. While single-objective alignment is well-studied, many real-world applications demand the simultaneous optimization of multiple conflicting rewards, e.g. optimizing both catalytic activity and specificity in protein engineering, or helpfulness and harmlessness for chatbots. Prior work has largely relied on linear reward scalarization, but this approach provably fails to recover non-convex regions of the Pareto front. In this paper, instead of scalarizing the rewards directly, we frame multi-objective RL itself as an optimization problem to be scalarized via smooth Tchebysheff scalarization, a recent technique that overcomes the shortcomings of linear scalarization. We use this formulation to derive Smooth Tchebysheff Optimization of Multi-Objective Preferences (STOMP), a novel offline RL algorithm that extends direct preference optimization to the multi-objective setting in a principled way by standardizing the individual rewards based on their observed distributions. We empirically validate STOMP on a range of protein engineering tasks by aligning three autoregressive protein language models on three laboratory datasets of protein fitness. Compared to state-of-the-art baselines, STOMP achieves the highest hypervolumes in eight of nine settings according to both offline off-policy and generative evaluations. We thus demonstrate that STOMP is a powerful, robust multi-objective alignment algorithm that can meaningfully improve post-trained models for multi-attribute protein optimization and beyond.

Related papers

• Joint Optimization of Training and Inference in Federated Edge Learning via Constrained Multi-Objective Deep Reinforcement Learning [9.37821062000313]
  Federated edge learning (FEEL) has emerged as a promising paradigm for achieving edge intelligence (EI)<n>We put forth an online optimization framework that jointly manages federated training and inference on resource-constrained edge devices.
  arXiv  Detail & Related papers  (2026-05-25T14:51:07Z)
• Large Language Models as Amortized Pareto-Front Generators for Constrained Bi-Objective Convex Optimization [21.83927813316836]
  DIPS is an end-to-end framework that fine-tunes large language models as amortized generators for constrained bi-objective convex optimization.<n>To make continuous optimization compatible with autoregressive language modeling, DIPS combines a compact discretization scheme, Numerically Grounded Token Initialization for new numerical tokens, and Three-Phase Curriculum Optimization.<n>With vLLM-accelerated inference, DIPS solves one instance in as little as 0.16 seconds and outperforms general-purpose and reasoning LLM baselines.
  arXiv  Detail & Related papers  (2026-05-12T13:20:57Z)
• MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization [56.074760766965085]
  Group-Relative Policy Optimization has emerged as an efficient paradigm for aligning Large Language Models (LLMs)<n>We propose MAESTRO, which treats reward scalarization as a dynamic latent policy, leveraging the model's terminal hidden states as a semantic bottleneck.<n>We formulate this as a contextual bandit problem within a bi-level optimization framework, where a lightweight Conductor network co-evolves with the policy by utilizing group-relative advantages as a meta-reward signal.
  arXiv  Detail & Related papers  (2026-01-12T05:02:48Z)
• Learning to Optimize Multi-Objective Alignment Through Dynamic Reward Weighting [48.87957020168614]
  Prior works in multi-reward learning typically use linear scalarization with fixed weights, which fail to capture effective online learning.<n>We introduce two approaches to increasing objective alignment, one for online learning, the other for space exploration.
  arXiv  Detail & Related papers  (2025-09-14T21:56:35Z)
• Linear Preference Optimization: Decoupled Gradient Control via Absolute Regularization [13.97375970293678]
  DPO (Direct Preference Optimization) has become a widely used offline preference optimization algorithm due to its simplicity and training stability.<n>We propose Linear Preference Optimization (LPO), a novel alignment framework featuring three key innovations.<n>First, we introduce gradient decoupling by replacing the log-sigmoid function with an absolute difference loss, thereby isolating the optimization dynamics.<n>Second, we improve stability through an offset constraint combined with a positive regularization term to preserve the chosen response quality.<n>Third, we implement controllable rejection suppression using gradient separation with straightforward estimation and a tunable coefficient that linearly regulates the descent of the rejection probability.
  arXiv  Detail & Related papers  (2025-08-20T10:17:29Z)
• Reparameterized LLM Training via Orthogonal Equivalence Transformation [54.80172809738605]
  We present POET, a novel training algorithm that uses Orthogonal Equivalence Transformation to optimize neurons.<n>POET can stably optimize the objective function with improved generalization.<n>We develop efficient approximations that make POET flexible and scalable for training large-scale neural networks.
  arXiv  Detail & Related papers  (2025-06-09T17:59:34Z)
• Self-Boost via Optimal Retraining: An Analysis via Approximate Message Passing [58.52119063742121]
  Retraining a model using its own predictions together with the original, potentially noisy labels is a well-known strategy for improving the model performance.<n>This paper addresses the question of how to optimally combine the model's predictions and the provided labels.<n>Our main contribution is the derivation of the Bayes optimal aggregator function to combine the current model's predictions and the given labels.
  arXiv  Detail & Related papers  (2025-05-21T07:16:44Z)
• Multi-Objective Preference Optimization: Improving Human Alignment of Generative Models [15.799929216215672]
  We introduce the Multi-Objective Preference Optimization (MOPO) algorithm, which frames alignment as a constrained KL-regularized optimization.<n>Unlike prior work, MOPO operates directly on pairwise preference data, requires no point-wise reward assumption, and avoids prompt-context engineering.
  arXiv  Detail & Related papers  (2025-05-16T05:58:26Z)
• Projection Optimization: A General Framework for Multi-Objective and Multi-Group RLHF [13.612504157832708]
  Reinforcement Learning with Human Feedback (RLHF) is a widely used fine-tuning approach that aligns machine learning model with human preferences.<n>In this work, we transform the non-linear aggregation problem into a series of sub-problems and extend our framework to handle multi-group scenarios.<n>We demonstrate that our algorithmic framework achieves sublinear regret and can be easily adapted to a reward-free algorithm.
  arXiv  Detail & Related papers  (2025-02-21T01:56:52Z)
• Training Deep Learning Models with Norm-Constrained LMOs [56.00317694850397]
  We propose a new family of algorithms that uses the linear minimization oracle (LMO) to adapt to the geometry of the problem.<n>We demonstrate significant speedups on nanoGPT training using our algorithm, Scion, without any reliance on Adam.
  arXiv  Detail & Related papers  (2025-02-11T13:10:34Z)
• Decoding-Time Language Model Alignment with Multiple Objectives [116.42095026960598]
  Existing methods primarily focus on optimizing LMs for a single reward function, limiting their adaptability to varied objectives. Here, we propose $textbfmulti-objective decoding (MOD)$, a decoding-time algorithm that outputs the next token from a linear combination of predictions. We show why existing approaches can be sub-optimal even in natural settings and obtain optimality guarantees for our method.
  arXiv  Detail & Related papers  (2024-06-27T02:46:30Z)
• Self-Exploring Language Models: Active Preference Elicitation for Online Alignment [88.56809269990625]
  We propose a bilevel objective optimistically biased towards potentially high-reward responses to actively explore out-of-distribution regions. Our experimental results demonstrate that when fine-tuned on Zephyr-7B-SFT and Llama-3-8B-Instruct models, Self-Exploring Language Models (SELM) significantly boosts the performance on instruction-following benchmarks.
  arXiv  Detail & Related papers  (2024-05-29T17:59:07Z)

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.