Decoding Open-Ended Information Seeking Goals from Eye Movements in Reading

Researchers have developed AI models that can decode readers' information-seeking goals solely from their eye movements while reading text. The study introduces new evaluation frameworks using large-scale eye tracking data and demonstrates success in both selecting correct goals from options and reconstructing precise goal formulations.

• →AI can now decode what readers are specifically looking for in a text by analyzing only their eye movements.
• →The research uses large-scale eye tracking data with hundreds of text-specific information seeking tasks in English.
• →Both discriminative and generative multimodal LLMs were developed and tested for goal decoding tasks.
• →The technology shows promise for educational and assistive applications that require real-time understanding of reader intentions.
• →This opens new scientific avenues for studying goal-driven reading behavior.