NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs

[2508.09473] NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs Skip to main content Search Submit Donate Log in Computer Science > Machine Learning arXiv:2508.09473 (cs) [Submitted on 13 Aug 2025 ( v1 ), last revised 25 Aug 2026 (this version, v2)] Title: NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs Authors: Birong Pan , Jianhao Chen , Mayi Xu , Qiankun Pi , Yuanyuan Zhu , Ming Zhong , Tieyun Qian View PDF HTML (experimental) Abstract: Ensuring robust safety alignment while preserving utility is critical for the reliable deployment of Large Language Models (LLMs). However, current techniques fundamentally suffer from intertwined deficiencies: insufficient robustness against malicious attacks, frequent refusal of benign queries, degradation in generated text quality and general task performance, the former two reflecting tensions in robust safety and the latter constituting utility impairment. We attribute these limitations to the coarse-grained layer-wise interventions in existing methods. To resolve this, we propose NeuronTune, a fine-grained framework that pinpoints and modulates sparse neurons to achieve simultaneous safety-utility optimization. Our approach first pinpoints safety-critical and utility-preserving neurons across all layers via attack-aware attribution, then adapts meta-learning to adaptively modulate their activations. Crucially, the intervention scope of NeuronTune is dynamically controlled via neuron inclusion thresholds, providing a flexible mechanism to prioritize either security-critical or utility-priority requirements. Extensive experimental results demonstrate that our method outperforms existing state-of-the-art technologies, achieving superior model safety while maintaining excellent utility. Comments: This work was accepted by WISE2026 Subjects: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2508.09473 [cs.LG] (or arXiv:2508.09473v2 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2508.09473 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Birong Pan [ view email ] [v1] Wed, 13 Aug 2025 04:05:28 UTC (555 KB) [v2] Tue, 25 Aug 2026 04:36:38 UTC (536 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.LG < prev | next > new | recent | 2025-08 Change to browse by: cs cs.AI cs.CL References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)

Related documents

• NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs#670367
• The diversity of lepidopteran spatial orientation strategies - neuronal mechanisms and emerging challenges in a changing world.#30764
• TMEM106B Acts as a Modifier of Cognitive and Motor Functions in Amyotrophic Lateral Sclerosis#476157
• Connectome-Based Modelling Reveals Orientation Maps in the Drosophila Optic Lobe#821277
• Controlling Refusal Behavior of LLMs via Stiefel-Constrained Rotation Steering#819023