Papers
arxiv:2602.02258

Alignment-Aware Model Adaptation via Feedback-Guided Optimization

Published on Feb 5
Authors:
,
,
,

Abstract

An alignment-aware fine-tuning framework uses policy-gradient-based regularization and adaptive gating to balance supervised and alignment-driven gradients, improving safety while maintaining task performance.

AI-generated summary

Fine-tuning is the primary mechanism for adapting foundation models to downstream tasks; however, standard approaches largely optimize task objectives in isolation and do not account for secondary yet critical alignment objectives (e.g., safety and hallucination avoidance). As a result, downstream fine-tuning can degrade alignment and fail to correct pre-existing misaligned behavior. We propose an alignment-aware fine-tuning framework that integrates feedback from an external alignment signal through policy-gradient-based regularization. Our method introduces an adaptive gating mechanism that dynamically balances supervised and alignment-driven gradients on a per-sample basis, prioritizing uncertain or misaligned cases while allowing well-aligned examples to follow standard supervised updates. The framework further learns abstention behavior for fully misaligned inputs, incorporating conservative responses directly into the fine-tuned model. Experiments on general and domain-specific instruction-tuning benchmarks demonstrate consistent reductions in harmful and hallucinated outputs without sacrificing downstream task performance. Additional analyses show robustness to adversarial fine-tuning, prompt-based attacks, and unsafe initializations, establishing adaptively gated alignment optimization as an effective approach for alignment-preserving and alignment-recovering model adaptation.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2602.02258
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2602.02258 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2602.02258 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2602.02258 in a Space README.md to link it from this page.

Collections including this paper 1