REVIEW 23 cited by
Safety Assessment of Chinese Large Language Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
With the rapid popularity of large language models such as ChatGPT and GPT-4, a growing amount of attention is paid to their safety concerns. These models may generate insulting and discriminatory content, reflect incorrect social values, and may be used for malicious purposes such as fraud and dissemination of misleading information. Evaluating and enhancing their safety is particularly essential for the wide application of large language models (LLMs). To further promote the safe deployment of LLMs, we develop a Chinese LLM safety assessment benchmark. Our benchmark explores the comprehensive safety performance of LLMs from two perspectives: 8 kinds of typical safety scenarios and 6 types of more challenging instruction attacks. Our benchmark is based on a straightforward process in which it provides the test prompts and evaluates the safety of the generated responses from the evaluated model. In evaluation, we utilize the LLM's strong evaluation ability and develop it as a safety evaluator by prompting. On top of this benchmark, we conduct safety assessments and analyze 15 LLMs including the OpenAI GPT series and other well-known Chinese LLMs, where we observe some interesting findings. For example, we find that instruction attacks are more likely to expose safety issues of all LLMs. Moreover, to promote the development and deployment of safe, responsible, and ethical AI, we publicly release SafetyPrompts including 100k augmented prompts and responses by LLMs.
Forward citations
Cited by 23 Pith papers
-
Paladin: Defending LLM-enabled Phishing Emails with a New Trigger-Tag Paradigm
A trigger-tag watermark embedded by fine-tuning lets modified LLMs mark their own phishing outputs for cheap detection.
-
CharacterBench: Benchmarking Character Customization of Large Language Models
CharacterBench provides a large Chinese-English benchmark and a fine-tuned judge model for measuring 11 dimensions of LLM character customization, with reported human-correlation improvements over GPT-4.
-
ELBench: A Multi-Dimensional Benchmark for Education-Facing Large Language Models
ELBench evaluates nine LLMs on capability, safety, teaching, and cultivation in one protocol, finding module profiles diverge, safety trades off with teaching, and all models share a style-over-fit judgment blind spot.
-
Calling the Bluff: Detecting Ever-Shifting Harmful Chat Dialogue via Ordered Reasoning Chain Regularization
BRACE detects harmful chat dialogue by regularizing a classifier with an ordered reasoning chain of topic, indicator, severity, and type, reaching 0.934 macro F1 on the authors' 9,000-dialogue test set.
-
EPT Benchmark: Evaluation of Persian Trustworthiness in Large Language Models
A new Persian-Islamic trustworthiness benchmark ranks Claude highest and Qwen lowest across eight LLMs and finds safety is the weakest dimension.
-
Libra: Large Chinese-based Safeguard for AI Content
Libra-Guard, a Chinese-language AI content safety model, reaches 86.8% accuracy on the new 5,720-sample Libra-Test benchmark, beating open-source baselines and approaching GPT-4o and Claude-3.5-Sonnet.
-
POEX: Towards Policy Executable Jailbreak Attacks Against the LLM-based Robots
POEX generates short adversarial suffixes that make LLM-based robots turn harmful instructions into executable robot policies, with about 60% average execution success across tested models.
-
Certifiable Safe RLHF: Semantic Grounding and Fixed Penalty Constraint Optimization for Safer LLM Alignment
A fixed ReLU penalty and a semantically labeled cost model are proposed to make RLHF safer, but the 'certifiable' guarantee is not fully supported.
-
SafeToolBench: Pioneering a Prospective Benchmark to Evaluating Tool Utilization Safety in LLMs
A new benchmark of 1,200 risky tool-use requests plus a nine-dimension scoring prompt claims to improve LLM safety, but the evaluation only includes risky samples, so high scores may just mean the model refuses everything.
-
GaussMaster: An LLM-based Database Copilot System
GaussMaster combines retrieval-augmented generation, safety controls, and diagnosis-tree-guided multi-agent orchestration to automate GaussDB Q&A and anomaly repair.
-
Xinyu AI Search: Enhanced Relevance and Comprehensive Results with Rich Answer Presentations
Xinyu, an integrated generative AI search engine with query decomposition, multi-source retrieval, and rich answer presentation, outperforms eight existing technologies in human evaluations.
-
Value Compass Benchmarks: A Platform for Fundamental and Validated Evaluation of LLMs Values
Value Compass Benchmarks is a live, self-evolving platform that scores 33 LLMs across 27 value dimensions from four value systems, aiming to reveal true behavioral alignment with human values.
-
RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting
A rationale-aware defensive prompting framework uses multimodal chain-of-thought and self-checking to reduce harmful MLLM outputs while preserving benign utility.
-
Trigger$^3$: Refining Query Correction via Adaptive Model Selector
Trigger3 uses three trained triggers to route Chinese search queries among a small correction model, an LLM, and the original query, improving F0.5 on two datasets while lowering LLM coverage.
-
Chinese SafetyQA: A Safety Short-form Factuality Benchmark for Large Language Models
A 2,000-question Chinese safety factuality benchmark shows most LLMs are inaccurate on safety knowledge, with retrieval helping more than self-reflection.
-
A Comprehensive Evaluation framework of Alignment Techniques for LLMs
The paper proposes a multi-dimensional framework to evaluate and compare LLM alignment techniques.
-
Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency
Fine-tuning small LLMs on benign data raises harmfulness scores, but those scores vary widely across random seeds, temperatures, and repeated runs, making single-run safety comparisons unreliable.
-
From LLMs to MLLMs to Agents: A Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystem
A structured survey of recent jailbreak attacks and defenses across LLMs, multimodal LLMs, and agents, with taxonomies for methods, datasets, metrics, and defenses.
-
The Tower of Babel Revisited: Multilingual Jailbreak Prompts on Closed-Source Large Language Models
A multilingual jailbreak benchmark on four proprietary LLMs finds language-dependent safety gaps and identifies a two-sided debate prompt as the most effective attack component.
-
A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy
A survey that organizes responsible-LLM research into five risk dimensions and four intervention phases, reviewing privacy, hallucination, value, toxicity, and jailbreak mitigation.
-
A Method for Enhancing the Safety of Large Model Generation Based on Multi-dimensional Attack and Defense
A multi-stage LLM-based attack/defense dataset pipeline improves reported safety scores of Llama-3.2-1B after SFT, but the evaluation is partly circular and lacks statistical baselines.
-
Navigating the Risks: A Survey of Security, Privacy, and Ethics Threats in LLM-Based Agents
A survey proposing a source-and-impact taxonomy (input, model, combined; security, privacy, ethics) for threats to LLM-based agents, with feature analysis and four case studies.
-
The Scales of Justitia: A Comprehensive Survey on Safety Evaluation of LLMs
A structured survey of LLM safety evaluation that proposes a why/what/where/how taxonomy and catalogs metrics, datasets, benchmarks, evaluators, and frameworks.
Discussion (0). Continue with ORCID to comment.