Dimensions overall score 5.0

• A Proactive Knowledge Behaviors Framework was proposed to capture human-human and human-AI knowledge seeking and provision behaviors.method 95%
• A Topic-Intent-Form taxonomy was developed to analyze climate-related data representing different knowledge behaviors.method 95%
• There is a substantial mismatch between current LLM benchmarks and real-world user needs for climate change information.result 95%
• Knowledge interaction patterns between humans and LLMs closely resemble those in human-human interactions.result 95%
• The study provides actionable guidance for benchmark design, RAG system development, and LLM training.result 95%
• Eight core climate-related datasets were used in the study, including LMSYS-Chat-1M, ClimateQ&A, Human-to-Human Questions (Reddit), and IPCCAR6.method 90%
• LLMs were used to filter out conversations unrelated to climate change from datasets like WildChat and LMSYS-Chat-1M.method 90%
• A Proactive Knowledge Behaviors Framework was proposed to capture human-human and human-AI knowledge seeking and provision behaviors.method 95%
• A Topic-Intent-Form taxonomy was developed to analyze climate-related data representing different knowledge behaviors.method 95%
• There is a substantial mismatch between current LLM benchmarks and real-world user needs for climate change information.result 95%
• Knowledge interaction patterns between humans and LLMs closely resemble those in human-human interactions.result 95%
• The study provides actionable guidance for benchmark design, RAG system development, and LLM training.result 95%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card