STAGE: Controlled Objective Admission for Multi-Preference LLM Alignment

• URL: http://arxiv.org/abs/2608.16553v1
• Date: Mon, 17 Aug 2026 13:24:11 GMT
• Title: STAGE: Controlled Objective Admission for Multi-Preference LLM Alignment
• Abstract summary: We propose a stability-guided active-set controller for controlled objective admission.<n>methodname starts from a small active set, retains admitted objectives, and expands when reward-deviation gates indicate low recent deviation or a patience budget is exhausted.
• Score: 7.5181381447764615
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-preference alignment is often framed as scalarization: combine reward dimensions, then optimize. This leaves a temporal decision underspecified: when should each preference dimension enter policy optimization? We propose \methodname, a stability-guided active-set controller for controlled objective admission. \methodname starts from a small active set, retains admitted objectives, and expands when reward-deviation gates indicate low recent deviation or a patience budget is exhausted. A probing phase estimates a hard-to-easy order, and adaptive weighting emphasizes underperforming active dimensions. Automatic evaluations with 15 training preferences and 16 held-out benchmark columns show that \methodname obtains higher averages than simultaneous scalarization and shared-budget adapted baselines. Component ablations and expansion dynamics further support cumulative retention, gated admission, and probing-derived ordering as useful design choices in this setting. These results position objective-entry timing as a concrete control variable in reward-vector RLHF.

Related papers

• Selective Ensemble Based on Preference-Directed Multi-Objective Bandits [90.75513823660775]
  We formalize a sequential decision problem under partially specified linear preferences.<n>We then propose a preference-directed upper confidence bound (PrefUCB) algorithm.<n> Experiments on large pre-trained model selective ensemble tasks and online asset allocation under institutional mandates validate our method.
  arXiv  Detail & Related papers  (2026-06-20T07:52:46Z)
• Dynamic Objective Selection with Safeguards and LLM Oversight for Financial Decision-Making [33.643275739813056]
  We propose DOSS (Dynamic Objective Selection with Safeguards), a learning-based selector that directly chooses the decision-relevant objective function at each time point.<n> DOSS formulates objective selection as a classification problem over objectives and performs sequential updates with a rolling window.<n>To mitigate misselection and excessive switching in deployment, DOSS applies confidence-aware gating with a fail-safe that overrides low-confidence proposals to a conservative default and enforces explicit controls tied to switching frequency.
  arXiv  Detail & Related papers  (2026-06-02T14:22:07Z)
• A Single Deep Preference-Conditioned Policy for Learning Pareto Coverage Sets [0.4369550829556578]
  We study the uniqueness and continuity of the preference-to-solution correspondence under nonlinear scalarization.<n>We use smooth Tchebycheff scalarization as a monotone utility to solve the problem.<n>We show that each update is equivalent to solving a Kullback-Leibler-regularized MDP with the previous policy as reference.
  arXiv  Detail & Related papers  (2026-05-09T13:35:50Z)
• Towards Disentangled Preference Optimization Dynamics Beyond Likelihood Displacement [33.80669933764735]
  We present a unified emphincentive-score decomposition of preference optimization.<n>We identify the emphdisentanglement band (DB), a simple, testable condition that characterizes when training can avoid likelihood displacement.<n>We propose a plug-and-play emphreward calibration (RC) that adaptively rebalances chosen versus rejected updates to satisfy the DB and likelihood displacement.
  arXiv  Detail & Related papers  (2026-04-20T13:23:27Z)
• Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization [60.87651283510059]
  Group Relative Policy Optimization (GRPO) effectively scales LLM reasoning but incurs prohibitive computational costs.<n>We propose Dynamic Pruning Policy Optimization (DPPO), a framework that enables dynamic pruning while preserving unbiased gradient estimation.<n>To mitigate the data sparsity induced by pruning, we introduce Dense Prompt Packing, a window-based greedy strategy.
  arXiv  Detail & Related papers  (2026-03-04T14:48:53Z)
• 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)
• Anchoring Values in Temporal and Group Dimensions for Flow Matching Model Alignment [61.80228667422234]
  VGPO redefines value estimation across both temporal and group dimensions.<n>It transforms the sparse terminal reward into dense, process-aware value estimates.<n>It replaces standard group normalization with a novel process enhanced by absolute values to maintain a stable optimization signal.
  arXiv  Detail & Related papers  (2025-12-13T16:31:26Z)
• Listwise Preference Diffusion Optimization for User Behavior Trajectories Prediction [41.53271688465831]
  We formulate User Behavior Trajectory Prediction (UBTP) as a new task setting that explicitly models long-term user preferences.<n>We introduce Listwise Preference Diffusion Optimization (LPDO), a diffusion-based training framework that directly optimize structured preferences over entire item sequences.<n>To rigorously evaluate multi-step prediction quality, we propose the task-specific metric Sequential Match (SeqMatch), which measures exact trajectory agreement, and adopt Perplexity (PPL), which assesses probabilistic fidelity.
  arXiv  Detail & Related papers  (2025-11-01T12:16:24Z)
• Multi-Preference Optimization: Generalizing DPO via Set-Level Contrasts [17.243429150450886]
  We propose $textbfMulti-Preference Optimization (MPO) to optimize over entire sets of responses.<n>MPO employs deviation-based weighting, which emphasizes outlier responses that deviate most from the mean reward.<n>We theoretically prove that MPO reduces alignment bias at a rate of $mathcalOleft(frac1sqrtnright)$ with respect to the number of responses per query.
  arXiv  Detail & Related papers  (2024-12-05T21:50:22Z)
• Dynamic Rewarding with Prompt Optimization Enables Tuning-free Self-Alignment of Language Models [54.381650481255235]
  We introduce a new tuning-free approach for self-alignment, Dynamic Rewarding with Prompt Optimization (O) Our approach leverages a search-based optimization framework that allows LLMs to iteratively self-improve and craft the optimal alignment instructions. Empirical evaluations on eight recent LLMs, both open and closed-sourced, demonstrate that DRPO significantly enhances alignment performance.
  arXiv  Detail & Related papers  (2024-11-13T16:15:38Z)
• Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment [103.12563033438715]
  Alignment in artificial intelligence pursues consistency between model responses and human preferences as well as values. Existing alignment techniques are mostly unidirectional, leading to suboptimal trade-offs and poor flexibility over various objectives. We introduce controllable preference optimization (CPO), which explicitly specifies preference scores for different objectives.
  arXiv  Detail & Related papers  (2024-02-29T12:12:30Z)

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.