paper-with-me

홈 › Papers

I Think, Therefore I am: Benchmarking Awareness of Large Language Models Using AwareBench

2024-01-31 · Yuan Li, Yue Huang, Yuli Lin, Siyuan Wu, Yao Wan, Lichao Sun

Do large language models (LLMs) exhibit any forms of awareness similar to humans? In this paper, we introduce AwareBench, a benchmark designed to evaluate awareness in LLMs. Drawing from theories in psychology and philosophy, we define awareness in LLMs as the ability to understand themselves as AI models and to exhibit social intelligence. Subsequently, we categorize awareness in LLMs into five dimensions, including capability, mission, emotion, culture, and perspective. Based on this taxonomy, we create a dataset called AwareEval, which contains binary, multiple-choice, and open-ended questions to assess LLMs' understandings of specific awareness dimensions. Our experiments, conducted on 13 LLMs, reveal that the majority of them struggle to fully recognize their capabilities and missions while demonstrating decent social intelligence. We conclude by connecting awareness of LLMs with AI alignment and safety, emphasizing its significance to the trustworthy and ethical development of LLMs. Our dataset and code are available at https://github.com/HowieHwong/Awareness-in-LLM.

📄 PDF Abstract BibTeX arXiv:2401.17882

Code (2)

howiehwong/awareness-in-llm 공식 구현
HowieHwong/TrustLLM

Tasks

BenchmarkingMultiple-choicePhilosophy

Similar Papers 제목 키워드 기반

Mind the Third Eye! Benchmarking Privacy Awareness in MLLM-powered Smartphone Agents

2025-08-27 · Zhixin Lin, Jungang Li, Shidong Pan, Yibo Shi 외 arxiv

Smartphones bring significant convenience to users but also enable devices to extensively record various types of personal information. Existing smartphone agents powered by Multimodal Large Language Models (MLLMs) have …

Rethinking Evaluation in the Era of Time Series Foundation Models: (Un)known Information Leakage Challenges

2025-10-15 · Marcel Meyer, Sascha Kaltenpoth, Kevin Zalipski, Oliver Müller arxiv

Time Series Foundation Models (TSFMs) represent a new paradigm for time-series forecasting, promising zero-shot predictions without the need for task-specific training or fine-tuning. However, similar to Large Language M…

Think-Reflect-Revise: A Policy-Guided Reflective Framework for Safety Alignment in Large Vision Language Models

2025-12-08 · Fenghua Weng, Chaochao Lu, Xia Hu, Wenqi Shao 외 arxiv

As multimodal reasoning improves the overall capabilities of Large Vision Language Models (LVLMs), recent studies have begun to explore safety-oriented reasoning, aiming to enhance safety awareness by analyzing potential…

Reinforcement LearningMultimodal Reasoning

Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs

2026-08-13 · Xinming Wang, Weinong Wang, Hongming Yang, Yansong Lin 외 arxiv

Hybrid-thinking multimodal large language models (MLLMs) allow a single model to alternate between deliberative thinking and latency-efficient non-thinking inference. Although these modes differ in reasoning budget, thei…

Reinforcement Learning

Thinking About Thinking: Evaluating Reasoning in Post-Trained Language Models

2025-10-18 · Pratham Singla, Shivank Garg, Ayush Singh, Ishan Garg 외 arxiv

Recent advances in post-training techniques have endowed Large Language Models (LLMs) with enhanced capabilities for tackling complex, logic-intensive tasks through the generation of supplementary planning tokens. This d…