Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Social Learning through Interactions with Other Agents: A Survey

Dylan Hillier, Cheston Tan, Jing Jiang

TL;DR

This survey synthesizes how social learning concepts—imitation, instructive guidance, and collaboration—appear in AI, with a focus on embodied agents and large language models (LLMs). It maps human social-learning paradigms to machine-learning techniques, highlighting imitation-based pretraining, demonstration-driven instruction, and collaborative multi-agent settings, including emergent LLM cooperation. Key contributions include a structured review of imitative, instructive, and collaborative approaches, analysis of feedback-driven alignment, and discussion of gaps in unifying socially embodied learning with permanent, multimodal adaptation. The work underscores the potential of LLM-enabled social learning to create more flexible, communicative agents while pointing to challenges in sample efficiency, theory of mind integration, and persistence of learning beyond initial social signals.

Abstract

Social learning plays an important role in the development of human intelligence. As children, we imitate our parents' speech patterns until we are able to produce sounds; we learn from them praising us and scolding us; and as adults, we learn by working with others. In this work, we survey the degree to which this paradigm -- social learning -- has been mirrored in machine learning. In particular, since learning socially requires interacting with others, we are interested in how embodied agents can and have utilised these techniques. This is especially in light of the degree to which recent advances in natural language processing (NLP) enable us to perform new forms of social learning. We look at how behavioural cloning and next-token prediction mirror human imitation, how learning from human feedback mirrors human education, and how we can go further to enable fully communicative agents that learn from each other. We find that while individual social learning techniques have been used successfully, there has been little unifying work showing how to bring them together into socially embodied agents.

Social Learning through Interactions with Other Agents: A Survey

TL;DR

This survey synthesizes how social learning concepts—imitation, instructive guidance, and collaboration—appear in AI, with a focus on embodied agents and large language models (LLMs). It maps human social-learning paradigms to machine-learning techniques, highlighting imitation-based pretraining, demonstration-driven instruction, and collaborative multi-agent settings, including emergent LLM cooperation. Key contributions include a structured review of imitative, instructive, and collaborative approaches, analysis of feedback-driven alignment, and discussion of gaps in unifying socially embodied learning with permanent, multimodal adaptation. The work underscores the potential of LLM-enabled social learning to create more flexible, communicative agents while pointing to challenges in sample efficiency, theory of mind integration, and persistence of learning beyond initial social signals.

Abstract

Social learning plays an important role in the development of human intelligence. As children, we imitate our parents' speech patterns until we are able to produce sounds; we learn from them praising us and scolding us; and as adults, we learn by working with others. In this work, we survey the degree to which this paradigm -- social learning -- has been mirrored in machine learning. In particular, since learning socially requires interacting with others, we are interested in how embodied agents can and have utilised these techniques. This is especially in light of the degree to which recent advances in natural language processing (NLP) enable us to perform new forms of social learning. We look at how behavioural cloning and next-token prediction mirror human imitation, how learning from human feedback mirrors human education, and how we can go further to enable fully communicative agents that learn from each other. We find that while individual social learning techniques have been used successfully, there has been little unifying work showing how to bring them together into socially embodied agents.
Paper Structure (30 sections, 2 figures, 1 table)

This paper contains 30 sections, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Social Learning in Humans
  3. Imitative Learning.
  4. Instructive Learning.
  5. Collaborative Learning.
  6. Social Learning in AI
  7. LLM Agents
  8. Objective of This Survey
  9. Imitative Learning
  10. LLM Pretraining
  11. Imitation Learning
  12. Section Summary:
  13. Instructive Learning
  14. Demonstration
  15. Feedback
  16. ...and 15 more sections

Figures (2)

  • Figure 1: The relationships among different social learning techniques, agents and the learning environment. Note: Imitation is initiated by the learner rather than the teacher. As in Figure \ref{['fig:section diagram']}, the learning approaches are coloured according to their classification in Tomasello's account of Social Learning.
  • Figure 2: The sections of this survey structured in relation to Tomasello's classification of Social Learning.