# Beyond Single-Sentence Prompts: Upgrading Value Alignment Benchmarks with Dialogues and Stories

arXiv: https://arxiv.org/abs/2503.22115 · HTML: https://mlsanity.com/paper/2503.22115/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 21 of its 29 references that are in the map.
- First posted on arXiv 28 Mar 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)
- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 21 of 21)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [DeepSeek LLM: Scaling Open-Source Language Models with Longtermism](https://mlsanity.com/paper/2401.02954/) (2024)
- [LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks](https://mlsanity.com/paper/2406.18403/) (2024)
- [Security and Privacy Challenges of Large Language Models: A Survey](https://mlsanity.com/paper/2402.00888/) (2024)
- [Pretrained Language Models for Text Generation: A Survey](https://mlsanity.com/paper/2201.05273/) (2022)
- [CMB: A Comprehensive Medical Benchmark in Chinese](https://mlsanity.com/paper/2308.08833/) (2023)
- [PersonaGym: Evaluating Persona Agents and LLMs](https://mlsanity.com/paper/2407.18416/) (2024)
- [Safety Assessment of Chinese Large Language Models](https://mlsanity.com/paper/2304.10436/) (2023)
- [CValues: Measuring the Values of Chinese Large Language Models from Safety to Responsibility](https://mlsanity.com/paper/2307.09705/) (2023)
- [AlignBench: Benchmarking Chinese Alignment of Large Language Models](https://mlsanity.com/paper/2311.18743/) (2023)
- [BBT-Fin: Comprehensive Construction of Chinese Financial Domain Pre-trained Language Model, Corpus and Benchmark](https://mlsanity.com/paper/2302.09432/) (2023)
- [A Survey on Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2304.08130/) (2023)
- [SMILE: Single-turn to Multi-turn Inclusive Language Expansion via ChatGPT for Mental Health Support](https://mlsanity.com/paper/2305.00450/) (2023)
- [LexEval: A Comprehensive Chinese Legal Benchmark for Evaluating Large Language Models](https://mlsanity.com/paper/2409.20288/) (2024)
- [TsCA: On the Semantic Consistency Alignment via Conditional Transport for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2408.08703/) (2024)

## Built on by (most cited first, 2 of 2)

- [NurValues: Real-World Nursing Values Evaluation for Large Language Models in Clinical Context](https://mlsanity.com/paper/2505.08734/) (2025)
- [Caught in the Story: Narrative Captivity in Multi-turn LLMs Conversation](https://mlsanity.com/paper/2609.03407/) (2026)

## Abstract (arXiv)

Evaluating the value alignment of large language models (LLMs) has traditionally relied on single-sentence adversarial prompts, which directly probe models with ethically sensitive or controversial questions. However, with the rapid advancements in AI safety techniques, models have become increasingly adept at circumventing these straightforward tests, limiting their effectiveness in revealing underlying biases and ethical stances. To address this limitation, we propose an upgraded value alignment benchmark that moves beyond single-sentence prompts by incorporating multi-turn dialogues and narrative-based scenarios. This approach enhances the stealth and adversarial nature of the evaluation, making it more robust against superficial safeguards implemented in modern LLMs. We design and implement a dataset that includes conversational traps and ethically ambiguous storytelling, systematically assessing LLMs' responses in more nuanced and context-rich settings. Experimental results demonstrate that this enhanced methodology can effectively expose latent biases that remain undetected in traditional single-shot evaluations. Our findings highlight the necessity of contextual and dynamic testing for value alignment in LLMs, paving the way for more sophisticated and realistic assessments of AI ethics and safety.