How to Enable Effective Cooperation Between Humans and NLP Models: A Survey of Principles, Formalizations, and Beyond

2025-01-10

AI Summary

This work addresses persistent cooperation barriers and evolutionary pathways in human–large language model (LLM) collaboration. Methodologically, it integrates game-theoretic modeling, cognitive science principles, multi-agent frameworks, and empirical behavioral analysis of LLMs to conduct a cross-paradigm comparative study. The contributions include: (1) the first unified taxonomy of human–model collaboration spanning the entire lifecycle, with a formally defined autonomous agent–based collaboration paradigm; (2) the construction of the inaugural human–model collaboration knowledge graph; and (3) the systematic identification of six fundamental open challenges—namely, interpretability, objective alignment, dynamic adaptability, trust calibration, value consistency, and cooperative agency emergence. Collectively, these results establish a rigorous theoretical foundation and methodological framework for designing trustworthy, robust, and goal-coherent human–LLM collaborative systems.

Technology Category

Humans and AI: Human-AI Collaboration / Human-AI TeamingMachine Learning: Large Multimodal Models (LMMs)Natural Language Processing: (Large) Language Models

Application Category

Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactionsUser Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendationEconomics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

With the advancement of large language models (LLMs), intelligent models have evolved from mere tools to autonomous agents with their own goals and strategies for cooperating with humans. This evolution has birthed a novel paradigm in NLP, i.e., human-model cooperation, that has yielded remarkable progress in numerous NLP tasks in recent years. In this paper, we take the first step to present a thorough review of human-model cooperation, exploring its principles, formalizations, and open challenges. In particular, we introduce a new taxonomy that provides a unified perspective to summarize existing approaches. Also, we discuss potential frontier areas and their corresponding challenges. We regard our work as an entry point, paving the way for more breakthrough research in this regard.

Problem

• Human-Robot Interaction
• Effective Cooperation
• Language Robots

Innovation

• Human-AI Collaboration
• Language Models
• Classification Framework