paper-with-me

홈 › Papers

Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs

2023-08-25 · Yuxia Wang, Haonan Li, Xudong Han, Preslav Nakov, Timothy Baldwin

With the rapid evolution of large language models (LLMs), new and hard-to-predict harmful capabilities are emerging. This requires developers to be able to identify risks through the evaluation of "dangerous capabilities" in order to responsibly deploy LLMs. In this work, we collect the first open-source dataset to evaluate safeguards in LLMs, and deploy safer open-source LLMs at a low cost. Our dataset is curated and filtered to consist only of instructions that responsible language models should not follow. We annotate and assess the responses of six popular LLMs to these instructions. Based on our annotation, we proceed to train several BERT-like classifiers, and find that these small classifiers can achieve results that are comparable with GPT-4 on automatic safety evaluation. Warning: this paper contains example data that may be offensive, harmful, or biased.

📄 PDF Abstract BibTeX arXiv:2308.13387

Code (1)

libr-ai/do-not-answer 공식 구현

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…

Similar Papers 제목 키워드 기반

On Evaluating the Durability of Safeguards for Open-Weight LLMs

2024-12-10 · Xiangyu Qi, Boyi Wei, Nicholas Carlini, Yangsibo Huang 외

Stakeholders -- from model developers to policymakers -- seek to minimize the dual-use risks of large language models (LLMs). An open challenge to this goal is whether technical safeguards can impede the misuse of LLMs, …

Safeguards Based on Copyable Context Cannot Provide Reliable Safety for LLMs

2026-07-30 · Pingyu Wu, Lingyao Zhu, Weiming Zhang, Nenghai Yu arxiv

Large language model safeguards decide whether to answer before seeing how an answer will be used. This creates a basic problem for dual-use tasks: the same answer can help an authorized professional or an attacker, whil…

A Chinese Dataset for Evaluating the Safeguards in Large Language Models

2024-02-19 · Yuxia Wang, Zenan Zhai, Haonan Li, Xudong Han 외

Many studies have demonstrated that large language models (LLMs) can produce harmful responses, exposing users to unexpected risks when LLMs are deployed. Previous studies have proposed comprehensive taxonomies of the ri…

The Safety Gap Toolkit: Evaluating Hidden Dangers of Open-Source Models

2025-07-08 · Ann-Kathrin Dombrowski, Dillon Bowen, Adam Gleave, Chris Cundy arxiv

Open-weight large language models (LLMs) unlock huge benefits in innovation, personalization, privacy, and democratization. However, their core advantage - modifiability - opens the door to systemic risks: bad actors can…

Evaluating the Promise and Pitfalls of LLMs in Hiring Decisions

2025-07-02 · Eitan Anzenberg, Arunava Samajpati, Sivasankaran Chandrasekar, Varun Kacholia

The use of large language models (LLMs) in hiring promises to streamline candidate screening, but it also raises serious concerns regarding accuracy and algorithmic bias where sufficient safeguards are not in place. In t…

Fairness