Safety-Utility Conflicts Are Not Global: Surgical Alignment via Head-Level Diagnosis (2026.acl-long)
Copied to clipboard
| Challenge: | Existing mitigation strategies rely on global gradient geometry to resolve alignment conflicts . however, they overlook Modular Heterogeneity within Transformers, resulting in suboptimal trade-offs . Conflict-Aware Sparse Tuning (CAST) combines head-level diagnosis with sparse fine-tuning . |
| Approach: | They propose a framework that integrates head-level diagnosis with sparse fine-tuning to address this limitation. |
| Outcome: | The proposed framework integrates head-level diagnosis with sparse fine-tuning to reduce alignment conflicts in LLMs. |
Similar Papers
SafeMERGE: Preserving Safety Alignment in Fine-Tuned Large Language Models via Selective Layer-Wise Model Merging (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent studies show that fine-tuning can erode safety alignment, causing LLMs to respond to harmful or unethical prompts. |
| Approach: | They propose a lightweight framework that restores safety while maintaining downstream performance. |
| Outcome: | The proposed framework reduces harmful outputs compared to other defenses, with negligible impact on utility. |
Preventing Safety Drift in Large Language Models via Coupled Weight and Activation Constraints (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing defenses constrain either weights or activations in isolation, without considering their coupled effects on safety. |
| Approach: | They propose a weight-activation constraint that enforces a precomputed safety subspace on weight updates and applies regularization to safety-critical features identified by sparse autoencoders. |
| Outcome: | The proposed model outperforms baselines even under high harmful data ratios. |
Safeguard Fine-Tuned LLMs Through Pre- and Post-Tuning Model Merging (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Fine-tuning large language models for downstream tasks often leads to catastrophic forgetting, notably degrading the safety of original alignments. |
| Approach: | They propose to merge the weights of pre- and post-fine-tuned models to improve safety while enhancing performance. |
| Outcome: | Experiments across different downstream tasks and models validate the method’s practicality and effectiveness. |
OASIS: Mitigating Harmful Fine-tuning Attacks on LLMs via Orthogonal and Adaptive Safety Alignment Strategy (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods to decouple safety enforcement from harmful feature acquisition rely on perturbation directions that conflict with harmful gradients . harmful fine-tuning attacks pose a significant challenge for service providers aiming to uphold rigorous safety standards. |
| Approach: | They propose an orthogonal and ad hoc safety alignment strategy to decouple safety enforcement from harmful feature acquisition. |
| Outcome: | Experiments on four large language models show that OASIS reduces the Harmful Score by 60% compared to baselines while maintaining stable task utility. |
LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion (2025.acl-long)
Copied to clipboard
| Challenge: | Existing safety alignment methods rely on fine-tuning, which inadvertently leads to the increased complexity and computational resources required. |
| Approach: | They propose a safety re-alignment framework with Low-Rank Safety Subspace Fusison that exploits low-rank safety characteristics of LLMs by constructing a low-ranked projection matrix to extract the principal components of safety vectors. |
| Outcome: | The proposed method exploits low-rank safety subspace of the LLMs and is stable during fine-tuning process and is isolated from the model’s general capabilities. |
The Unintended Trade-off of AI Alignment: Balancing Hallucination Mitigation and Safety in LLMs (2026.findings-eacl)
Copied to clipboard
| Challenge: | Hallucination in large language models has been studied, but a side effect remains unrecognized . a new study examines the trade-off between truthfulness and safety alignment . |
| Approach: | They propose a method that disentangles hallucination from hallucinian features using sparse autoencoders. |
| Outcome: | The proposed method preserves refusal behavior and task utility while maintaining safety alignment. |
Assessing and Mitigating Medical Knowledge Drift and Conflicts in Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Rapid medical concept drift can lead LLMs to provide incorrect or outdated advice. |
| Approach: | They propose to evaluate how large language models manage knowledge conflicts in clinical guidelines. |
| Outcome: | The proposed benchmark evaluates how LLMs manage varied knowledge conflicts in clinical guidelines. |
The Art of (Mis)alignment: How Fine-Tuning Methods Effectively Misalign and Realign LLMs in Post-Training (2026.findings-acl)
Copied to clipboard
Rui Zhang, Hongwei Li, Yun Shen, Xinyue Shen, Wenbo Jiang, Guowen Xu, Yang Liu, Michael Backes, Yang Zhang
| Challenge: | Misaligned large language models can magnify harm by exploiting them to undermine safety . et al., 2022b; Bai e.t., 2023): misalignment, realignment and model-specific resistance are important . |
| Approach: | They evaluate four methods to identify a mechanism asymmetry between attack and defense . they find that ORPO is most effective for misalignment, but DPO excels in realignment . |
| Outcome: | The proposed methods show a mechanism asymmetry between attack and defense . the proposed methods excel in realignment, but at the expense of model utility . |
Combining Domain and Alignment Vectors Provides Better Knowledge-Safety Trade-offs in LLMs (2025.acl-short)
Copied to clipboard
| Challenge: | Large language models (LLMs) excel in specific technical fields, but are not explicitly trained to be safe. |
| Approach: | They propose a model merging-based alignment method that allows for safer domain-specific models that preserve their utility. |
| Outcome: | The proposed method improves safety alignment on LLMs with minimal degradation on domain-specific benchmarks. |
LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods for fine-tuning large language models for specialized tasks are costly and time-consuming. |
| Approach: | They propose a framework that locates task-specific neurons via gradient-based attribution and dynamically Elects critical neurons through multi-model importance fusion. |
| Outcome: | The proposed framework reduces harmful response rates while preserving 95% of utility performance. |