Backtracking Improves Generation Safety

• URL: http://arxiv.org/abs/2409.14586v1
• Date: Sun, 22 Sep 2024 20:28:40 GMT
• Title: Backtracking Improves Generation Safety
• Authors: Yiming Zhang, Jianfeng Chi, Hailey Nguyen, Kartikeya Upasani, Daniel M. Bikel, Jason Weston, Eric Michael Smith,
• Abstract summary: We propose backtracking, a technique that allows language models to "undo" and recover from their own unsafe generation. We show that models trained to backtrack are consistently safer than baseline models.
• Score: 27.214507875399086
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Text generation has a fundamental limitation almost by definition: there is no taking back tokens that have been generated, even when they are clearly problematic. In the context of language model safety, when a partial unsafe generation is produced, language models by their nature tend to happily keep on generating similarly unsafe additional text. This is in fact how safety alignment of frontier models gets circumvented in the wild, despite great efforts in improving their safety. Deviating from the paradigm of approaching safety alignment as prevention (decreasing the probability of harmful responses), we propose backtracking, a technique that allows language models to "undo" and recover from their own unsafe generation through the introduction of a special [RESET] token. Our method can be incorporated into either SFT or DPO training to optimize helpfulness and harmlessness. We show that models trained to backtrack are consistently safer than baseline models: backtracking Llama-3-8B is four times more safe than the baseline model (6.1\% $\to$ 1.5\%) in our evaluations without regression in helpfulness. Our method additionally provides protection against four adversarial attacks including an adaptive attack, despite not being trained to do so.

Related papers

• Safety Pretraining: Toward the Next Generation of Safe AI [61.2816320807586]
  We present a data-centric pretraining framework that builds safety into the model from the start. Our contributions include: (i) a safety classifier trained on 10,000 GPT-4 labeled examples, used to filter 600B tokens; (ii) the largest synthetic safety dataset to date, generated via recontextualization of harmful web data; and (iv) Harmfulness-Tag annotations injected during pretraining to flag unsafe content.
  arXiv  Detail & Related papers  (2025-04-23T17:58:08Z)
• Safe Vision-Language Models via Unsafe Weights Manipulation [75.04426753720551]
  We revise safety evaluation by introducing Safe-Ground, a new set of metrics that evaluate safety at different levels of granularity. We take a different direction and explore whether it is possible to make a model safer without training, introducing Unsafe Weights Manipulation (UWM) UWM uses a calibration set of safe and unsafe instances to compare activations between safe and unsafe content, identifying the most important parameters for processing the latter.
  arXiv  Detail & Related papers  (2025-03-14T17:00:22Z)
• Backtracking for Safety [11.141166381133054]
  Large language models (LLMs) have demonstrated remarkable capabilities across various tasks, but ensuring their safety and alignment with human values remains crucial. Current safety alignment methods, such as supervised fine-tuning and reinforcement learning-based approaches, can exhibit vulnerabilities to adversarial attacks. We propose a novel backtracking method designed to address these limitations.
  arXiv  Detail & Related papers  (2025-03-11T22:04:22Z)
• Vulnerability Mitigation for Safety-Aligned Language Models via Debiasing [12.986006070964772]
  Safety alignment is an essential research topic for real-world AI applications. Our study first identified the difficulty of eliminating such vulnerabilities without sacrificing the model's helpfulness. Our method could enhance the model's helpfulness while maintaining safety, thus improving the trade-off-front.
  arXiv  Detail & Related papers  (2025-02-04T09:31:54Z)
• Evaluating Defences against Unsafe Feedback in RLHF [26.872318173182414]
  This paper looks at learning from unsafe feedback with reinforcement learning. We find that safety-aligned LLMs easily explore unsafe action spaces via generating harmful text. In order to protect against this vulnerability, we adapt a number of both "implict" and "explicit" harmful fine-tuning defences.
  arXiv  Detail & Related papers  (2024-09-19T17:10:34Z)
• What Makes and Breaks Safety Fine-tuning? A Mechanistic Study [64.9691741899956]
  Safety fine-tuning helps align Large Language Models (LLMs) with human preferences for their safe deployment. We design a synthetic data generation framework that captures salient aspects of an unsafe input. Using this, we investigate three well-known safety fine-tuning methods.
  arXiv  Detail & Related papers  (2024-07-14T16:12:57Z)
• Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training [67.30423823744506]
  This study addresses a critical gap in safety tuning practices for Large Language Models (LLMs) We introduce a novel approach, Decoupled Refusal Training (DeRTa), designed to empower LLMs to refuse compliance to harmful prompts at any response position. DeRTa incorporates two novel components: (1) Maximum Likelihood Estimation with Harmful Response Prefix, which trains models to recognize and avoid unsafe content by appending a segment of harmful response to the beginning of a safe response, and (2) Reinforced Transition Optimization (RTO), which equips models with the ability to transition from potential harm to safety refusal consistently throughout the harmful
  arXiv  Detail & Related papers  (2024-07-12T09:36:33Z)
• Single Character Perturbations Break LLM Alignment [20.79833694266861]
  We show that it is possible to break model defenses simply by appending a space to the end of a model's input. We examine the causes of this behavior, finding that the contexts in which single spaces occur in tokenized training data encourage models to generate lists when prompted. Our findings underscore the fragile state of current model alignment and promote the importance of developing more robust alignment methods.
  arXiv  Detail & Related papers  (2024-07-03T16:03:10Z)
• SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance [48.80398992974831]
  SafeAligner is a methodology implemented at the decoding stage to fortify defenses against jailbreak attacks. We develop two specialized models: the Sentinel Model, which is trained to foster safety, and the Intruder Model, designed to generate riskier responses. We show that SafeAligner can increase the likelihood of beneficial tokens, while reducing the occurrence of harmful ones.
  arXiv  Detail & Related papers  (2024-06-26T07:15:44Z)
• Emulated Disalignment: Safety Alignment for Large Language Models May Backfire! [65.06450319194454]
  Large language models (LLMs) undergo safety alignment to ensure safe conversations with humans. This paper introduces a training-free attack method capable of reversing safety alignment. We name this method emulated disalignment (ED) because sampling from this contrastive distribution provably emulates the result of fine-tuning to minimize a safety reward.
  arXiv  Detail & Related papers  (2024-02-19T18:16:51Z)
• Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! [88.90694413503614]
  We find that the safety alignment of LLMs can be compromised by fine-tuning. We jailbreak GPT-3.5 Turbo's safety guardrails by fine-tuning it on only 10 such examples. We advocate for further research efforts toward reinforcing safety protocols for the custom fine-tuning of aligned LLMs.
  arXiv  Detail & Related papers  (2023-10-05T17:12:17Z)
• Constructing Highly Inductive Contexts for Dialogue Safety through Controllable Reverse Generation [65.48908724440047]
  We propose a method called emphreverse generation to construct adversarial contexts conditioned on a given response. We test three popular pretrained dialogue models (Blender, DialoGPT, and Plato2) and find that BAD+ can largely expose their safety problems.
  arXiv  Detail & Related papers  (2022-12-04T12:23:41Z)

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.