| Challenge: | Excessive safety can lead to over-refusal, where models reject harmful-looking yet benign queries, severely limiting utility. |
| Approach: | They propose a lightweight training-based approach that reshapes the distributions of harmful and benign samples within the model’s decision space by using a single-token prefix. |
| Outcome: | The proposed approach can distinguish between harmful and benign samples while keeping the model frozen. |
Similar Papers
PURE: Aligning LLM via Pluggable Query Reformulation for Enhanced Helpfulness (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) depend on vast amounts of text data sourced from the Internet for their training. |
| Approach: | They propose a new alignment paradigm that reformulates risky queries into highly relevant yet harmless ones before feeding them into LLMs. |
| Outcome: | The proposed approach eliminates the high costs of training base LLMs and achieves a promising balance of harmlessness and helpfulness. |
Safeguarding LLM Fine-tuning via Push-Pull Distributional Alignment (2026.acl-long)
Copied to clipboard
| Challenge: | Existing safety defenses for large language models fail to explicitly repel harmful patterns . Optimal transport (SOT) allows for safe fine-tuning without sacrificing safety . |
| Approach: | They propose a framework that reframes safe fine-tuning from instance-level filtering challenge to distribution-level alignment task grounded in Optimal Transport. |
| Outcome: | a new framework improves safety of large language models while maintaining competitive performance . the proposed framework reduces the risk of errors and improves model performance compared to baselines . |
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. |
Aligning Large Language Models via Fine-grained Supervision (2024.acl-short)
Copied to clipboard
| Challenge: | Pre-trained large-scale language models often generate biased or toxic text, misaligning with human intentions. |
| Approach: | They propose to use human feedback to improve LLM alignment by fine-grained token supervision . they ask annotators to edit less preferred responses to make them more favorable . |
| Outcome: | The proposed method improves LLM alignment by up to 5.1% in terms of win rate compared with the traditional model. |
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. |
Separate the Wheat from the Chaff: A Post-Hoc Approach to Safety Re-Alignment for Fine-Tuned Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models achieve effective safety alignment at the time of release, but fine-tuning often compromises safety mechanisms. |
| Approach: | They propose a method that performs safety realignment for large language models . they identify unsafe delta parameters from the fine-tuned models and recalibrate the retained parameters . |
| Outcome: | The proposed method improves safety performance on safety benchmarks and jailbreak attacks while maintaining their performance on downstream tasks. |
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. |
SLM as Guardian: Pioneering AI Safety with Small Language Model (2024.emnlp-industry)
Copied to clipboard
Ohjoon Kwon, Donghyeon Jeon, Nayoung Choi, Gyu-Hwung Cho, Hwiyeol Jo, Changbong Kim, Hyunwoo Lee, Inho Kang, Sun Kim, Taiwoo Park
| Challenge: | Prior safety research on large language models focused on aligning them to safety requirements, but internalizing such safeguard features into larger models brought challenges of higher training cost and unintended degradation of helpfulness. |
| Approach: | They propose a multi-task learning mechanism that integrates harmful query detection and safeguard response into a single model. |
| Outcome: | The proposed approach outperforms the publicly available LLMs in harmful query detection and safeguard response generation. |
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. |
Please refuse to answer me! Mitigating Over-Refusal in Large Language Models via Adaptive Contrastive Decoding (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for mitigating over-refusal can't maintain low refusal ratio for harmless queries while keeping high for malicious queries. |
| Approach: | They propose a model-agnostic approach to mitigate over-refusal in large language models . they propose an adaptive contrastive decoding strategy that incorporates or removes the refusal token distribution . |
| Outcome: | The proposed approach reduces the refusal ratio for over-refusal queries by 10.35% while increasing the refusal rate for malicious queries by 0.13%. |