The Realignment Problem: When Right becomes Wrong in LLMs

• URL: http://arxiv.org/abs/2511.02623v1
• Date: Tue, 04 Nov 2025 14:52:58 GMT
• Title: The Realignment Problem: When Right becomes Wrong in LLMs
• Authors: Aakash Sen Sharma, Debdeep Sanyal, Vivek Srivastava, Shirish Karande, Murari Mandal,
• Abstract summary: The alignment of Large Language Models with human values is central to their safe deployment, yet current models fail to keep pace with evolving norms and policies.<n>Existing unlearning methods act as blunt instruments that erode utility rather than enable precise policy updates.<n>We introduce TRACE, a framework for principled unlearning that reconceives realignment as a programmatic policy problem.
• Score: 6.8304813545377
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: The alignment of Large Language Models (LLMs) with human values is central to their safe deployment, yet current practice produces static, brittle, and costly-to-maintain models that fail to keep pace with evolving norms and policies. This misalignment, which we term the Alignment-Reality Gap, poses a growing challenge for reliable long-term use. Existing remedies are inadequate: large-scale re-annotation is economically prohibitive, and standard unlearning methods act as blunt instruments that erode utility rather than enable precise policy updates. We introduce TRACE (Triage and Re-align by Alignment Conflict Evaluation), a framework for principled unlearning that reconceives re-alignment as a programmatic policy application problem. TRACE programmatically triages existing preference data against a new policy, identifies high-impact conflicts via a alignment impact score, and applies a hybrid optimization that cleanly inverts, discards, or preserves preferences while safeguarding model performance. Empirical results show that TRACE achieves robust re-alignment across diverse model families (Qwen2.5-7B, Gemma-2-9B, Llama-3.1-8B). On both synthetic benchmarks and the PKU-SafeRLHF dataset under complex policy shift, TRACE enforces new principles without degrading general capabilities. Our work establishes a scalable, dynamic, and cost-effective paradigm for maintaining LLM alignment, providing a foundation for sustainable and responsible AI deployment.

Related papers

• Unifying Stable Optimization and Reference Regularization in RLHF [64.16830602324345]
  This paper introduces a unified regularization approach that balances objectives of preventing reward hacking and maintaining stable policy updates.<n>Our simple yet principled alignment objective yields a weighted supervised fine-tuning loss with a superior trade-off, which demonstrably improves both alignment results and implementation complexity.
  arXiv  Detail & Related papers  (2026-02-12T03:31:19Z)
• Not All Preferences Are Created Equal: Stability-Aware and Gradient-Efficient Alignment for Reasoning Models [52.48582333951919]
  We propose a dynamic framework designed to enhance alignment reliability by maximizing the Signal-to-Noise Ratio of policy updates.<n>SAGE (Stability-Aware Gradient Efficiency) integrates a coarse-grained curriculum mechanism that refreshes candidate pools based on model competence.<n> Experiments on multiple mathematical reasoning benchmarks demonstrate that SAGE significantly accelerates convergence and outperforms static baselines.
  arXiv  Detail & Related papers  (2026-02-01T12:56:10Z)
• 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)
• BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping [69.74252624161652]
  We propose BAlanced Policy Optimization with Adaptive Clipping (BAPO)<n>BAPO dynamically adjusts clipping bounds to adaptively re-balance positive and negative contributions, preserve entropy, and stabilize RL optimization.<n>On AIME 2024 and AIME 2025 benchmarks, our 7B BAPO model surpasses open-source counterparts such as SkyWork-OR1-7B.
  arXiv  Detail & Related papers  (2025-10-21T12:55:04Z)
• ADARL: Adaptive Low-Rank Structures for Robust Policy Learning under Uncertainty [28.291179179647795]
  We propose textbfAdaptive Rank Representation (AdaRL), a bi-level optimization framework that improves robustness.<n>At the lower level, AdaRL performs policy optimization under fixed-rank constraints with dynamics sampled from a Wasserstein ball around a centroid model.<n>At the upper level, it adaptively adjusts the rank to balance the bias--variance trade-off, projecting policy parameters onto a low-rank manifold.
  arXiv  Detail & Related papers  (2025-10-13T20:05:34Z)
• Stabilizing Policy Gradients for Sample-Efficient Reinforcement Learning in LLM Reasoning [77.92320830700797]
  Reinforcement Learning has played a central role in enabling reasoning capabilities of Large Language Models.<n>We propose a tractable computational framework that tracks and leverages curvature information during policy updates.<n>The algorithm, Curvature-Aware Policy Optimization (CAPO), identifies samples that contribute to unstable updates and masks them out.
  arXiv  Detail & Related papers  (2025-10-01T12:29:32Z)
• Steerable Adversarial Scenario Generation through Test-Time Preference Alignment [58.37104890690234]
  Adversarial scenario generation is a cost-effective approach for safety assessment of autonomous driving systems.<n>We introduce a new framework named textbfSteerable textbfAdversarial scenario textbfGEnerator (SAGE)<n>SAGE enables fine-grained test-time control over the trade-off between adversariality and realism without any retraining.
  arXiv  Detail & Related papers  (2025-09-24T13:27:35Z)
• Dual Alignment Maximin Optimization for Offline Model-based RL [37.563771012143725]
  offline reinforcement agents face significant deployment challenges due to the synthetic-to-real distribution mismatch.<n>In this paper, we first shift the focus from model reliability to policy discrepancies while optimizing for expected returns, and then self-consistently incorporate synthetic data.<n>It is a unified framework to ensure both model-environment policy consistency and synthetic and data offline.
  arXiv  Detail & Related papers  (2025-02-02T16:47:35Z)
• Robust Offline Reinforcement Learning with Linearly Structured f-Divergence Regularization [11.739526562075339]
  The Robust Regularized Markov Decision Process (RRMDP) is proposed to learn policies robust to dynamics shifts by adding regularization to the transition dynamics in the value function.<n>We develop the Robust Regularized Pessimistic Value Iteration (R2PVI) algorithm that employs linear function approximation for robust policy learning in $d$-RRMDPs with $f$-divergence based regularization terms on transition kernels.
  arXiv  Detail & Related papers  (2024-11-27T18:57:03Z)
• PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback [106.63518036538163]
  We present a novel unified bilevel optimization-based framework, textsfPARL, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning. Our framework addressed these concerns by explicitly parameterizing the distribution of the upper alignment objective (reward design) by the lower optimal variable. Our empirical results substantiate that the proposed textsfPARL can address the alignment concerns in RL by showing significant improvements.
  arXiv  Detail & Related papers  (2023-08-03T18:03:44Z)
• A Regularized Implicit Policy for Offline Reinforcement Learning [54.7427227775581]
  offline reinforcement learning enables learning from a fixed dataset, without further interactions with the environment. We propose a framework that supports learning a flexible yet well-regularized fully-implicit policy. Experiments and ablation study on the D4RL dataset validate our framework and the effectiveness of our algorithmic designs.
  arXiv  Detail & Related papers  (2022-02-19T20:22:04Z)

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.