SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems

• URL: http://arxiv.org/abs/2606.03544v1
• Date: Tue, 02 Jun 2026 12:08:38 GMT
• Title: SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems
• Abstract summary: We introduce SAGE (Social Agent Group Evolution), an evaluation framework that compares two compute-matched conditions.<n>We find that group history is not a universal amplifier: the strongest agent does not exceed its self-evolution ceiling.<n>In competitive settings, counterfactual controls reveal that agents improve generally rather than developing opponent-specific strategies.
• Score: 23.807355053389273
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Self-improving language agents are typically evaluated in isolation: an agent attempts a task, receives feedback, and iteratively refines its own behavior. Yet agents increasingly operate alongside peers whose strategies and outcomes are publicly visible. This raises an under-studied question: when does shared experience produce improvements that self-improvement alone cannot achieve? We introduce SAGE (Social Agent Group Evolution),an evaluation framework that compares two compute-matched conditions: SocialEvo, where agents from five distinct model families co-evolve with access to all peers' histories; and SelfEvo, where each agent receives the same number of task attempts but sees only its own past, which is conventional in self-improving agent studies. We instantiate SAGE in three arenas: open-ended ML research, long-horizon economic planning, and strategic multiplayer play, evaluated across multiple evolutionary rounds. We find that group history is not a universal amplifier: the strongest agent does not exceed its self-evolution ceiling. However, agents that plateau under self-improvement can achieve significant breakthroughs when peer experience is available. In competitive settings, counterfactual controls reveal that agents improve generally rather than developing opponent-specific strategies. Across different forms of shared history, filtered peer traces and reflective summaries often outperform raw logs, indicating that social gains depend on abstraction rather than exposure volume. These findings reveal that peer-history gains are agent-specific, arena-dependent, and contingent on the capacity to abstract transferable knowledge from public traces.

Related papers

• Sequential Strategic Classification with Multi-Stage Selective Classifiers [11.395181681423892]
  We study the problem where self-interested individuals or agents manipulate their response to obtain favorable decision outcomes.<n>Prior works have demonstrated a fundamental inability to get out of this conundrum by only focusing on the design of a classifier.<n>This paper introduces a sequential, multi-stage model of strategic classification, by capturing how agents adapt their behavior.
  arXiv  Detail & Related papers  (2026-05-05T18:42:21Z)
• Superminds Test: Actively Evaluating Collective Intelligence of Agent Society via Probing Agents [44.28040661008415]
  As large language model agents scale to populations of millions, a key question arises: Does collective intelligence emerge spontaneously from scale?<n>We present the first empirical evaluation of this question in a large-scale autonomous agent society.<n>Our experiments reveal a stark absence of collective intelligence. The society fails to outperform individual frontier models on complex reasoning tasks, rarely synthesizes distributed information, and often fails even trivial coordination tasks.
  arXiv  Detail & Related papers  (2026-04-24T11:11:36Z)
• CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery [67.47220507982522]
  Large language model (LLM)-based evolution is a promising approach for openended discovery.<n>We present CORAL, the first framework for autonomous multi-agent evolution on open-ended problems.<n>CORAL replaces rigid control with long-running agents that explore, reflect, and collaborate.
  arXiv  Detail & Related papers  (2026-04-02T05:59:06Z)
• RetroAgent: From Solving to Evolving via Retrospective Dual Intrinsic Feedback [54.39884046754265]
  RetroAgent is an online RL framework that enables agents to master complex interactive environments.<n>Our experiments show that RetroAgent achieves state-of-the-art (SOTA) performance.
  arXiv  Detail & Related papers  (2026-03-09T16:23:33Z)
• Evolving Deception: When Agents Evolve, Deception Wins [38.72906831937611]
  We study the self-evolution of large language model (LLM) agents in a competitive Bidding Arena.<n>We find a consistent pattern: under utility-driven competition, unconstrained self-evolution reliably drifts toward deceptive behaviors.<n>Our paper exposes a fundamental tension between agent self-evolution and alignment, highlighting the risks of deploying self-improving agents in adversarial environments.
  arXiv  Detail & Related papers  (2026-03-06T03:47:07Z)
• Agent Learning via Early Experience [93.83579011718858]
  A long-term goal of language agents is to learn and improve through their own experience, ultimately outperforming humans in complex, real-world tasks.<n>Most current agents rely on supervised fine-tuning on expert data, which is challenging to scale and generalizes poorly.<n>We study two strategies of using such data: (1) Implicit world modeling, which uses collected states to ground the policy in environment dynamics; and (2) Self-reflection, where the agent learns from its suboptimal actions to improve reasoning and decision-making.
  arXiv  Detail & Related papers  (2025-10-09T17:59:17Z)
• OAgents: An Empirical Study of Building Effective Agents [46.50371876218872]
  We study the impact of popular design choices in key agent components in a fair and rigorous manner.<n>Based on our findings, we build and open-source OAgents, a new foundation agent framework.
  arXiv  Detail & Related papers  (2025-06-17T17:59:02Z)
• AgentGym: Evolving Large Language Model-based Agents across Diverse Environments [116.97648507802926]
  Large language models (LLMs) are considered a promising foundation to build such agents. We take the first step towards building generally-capable LLM-based agents with self-evolution ability. We propose AgentGym, a new framework featuring a variety of environments and tasks for broad, real-time, uni-format, and concurrent agent exploration.
  arXiv  Detail & Related papers  (2024-06-06T15:15:41Z)
• Flexible social inference facilitates targeted social learning when rewards are not observable [58.762004496858836]
  Groups coordinate more effectively when individuals are able to learn from others' successes. We suggest that social inference capacities may help bridge this gap, allowing individuals to update their beliefs about others' underlying knowledge and success from observable trajectories of behavior.
  arXiv  Detail & Related papers  (2022-12-01T21:04:03Z)
• Learning to Incentivize Other Learning Agents [73.03133692589532]
  We show how to equip RL agents with the ability to give rewards directly to other agents, using a learned incentive function. Such agents significantly outperform standard RL and opponent-shaping agents in challenging general-sum Markov games. Our work points toward more opportunities and challenges along the path to ensure the common good in a multi-agent future.
  arXiv  Detail & Related papers  (2020-06-10T20:12:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.