Collective Alignment in LLM Multi-Agent Systems: Disentangling Bias from Cooperation via Statistical Physics

• URL: http://arxiv.org/abs/2605.10528v1
• Date: Mon, 11 May 2026 13:13:44 GMT
• Title: Collective Alignment in LLM Multi-Agent Systems: Disentangling Bias from Cooperation via Statistical Physics
• Abstract summary: We present a model-agnostic statistical-physics method to disentangle social conformity from intrinsic bias, compute critical exponents, and probe the collective behavior and possible phase transitions of multi-agent systems.<n>We find that collective alignment is dominated by an intrinsic bias rather than by cooperative neighbor coupling, producing field-driven crossovers instead of genuine phase transitions.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We investigate the emergent collective dynamics of LLM-based multi-agent systems on a 2D square lattice and present a model-agnostic statistical-physics method to disentangle social conformity from intrinsic bias, compute critical exponents, and probe the collective behavior and possible phase transitions of multi-agent systems. In our framework, each node of an $L\!\times\!L$ lattice hosts an identical LLM agent holding a binary state ($+1$/$-1$, mapped to yes/no) and updating it by querying the model conditioned on the four nearest-neighbor states. The sampler temperature $T$ serves as the sole control parameter. Across three open-weight models (llama3.1:8b, phi4-mini:3.8b, mistral:7b), we measure magnetization and susceptibility under a global-flip protocol designed to probe $\mathbb{Z}_2$ symmetry. All models display temperature-driven order-disorder crossovers and susceptibility peaks; finite-size scaling on even-$L$ lattices yields effective exponents $γ/ν$ whose values are model-dependent, close to but incompatible with the 2D Ising universality class ($γ/ν=7/4$). Our method enables the extraction of effective $β$-weighted couplings $\tilde{J}(T)$ and fields $\tilde{h}(T)$, which serve as a measure of social conformity and intrinsic bias. In the models we analyzed, we found that collective alignment is dominated by an intrinsic bias ($\tilde{h}\gg\tilde{J}$) rather than by cooperative neighbor coupling, producing field-driven crossovers instead of genuine phase transitions. These effective parameters vary qualitatively across models, providing compact collective-behavior fingerprints for LLM agents and a quantitative diagnostic for the reliability of multi-agent consensus and collective alignment.

Related papers

• World models of environment, agent and joint agent-environment systems [76.81951414842736]
  We argue that there is a prior distinction: which channel they model.<n>We consider three cases: the environment channel $O_: mid A_:$, the agent channel $A_: mid O_:$, and the realised joint process $(A, O)_:$, equivalently viewed as a channel with no inputs.
  arXiv  Detail & Related papers  (2026-07-23T13:35:03Z)
• Inverting the Bellman Equation: From $Q$-Values to World Models [57.827849584133425]
  We prove that value-based agents trained on a sufficiently rich set of reward functions implicitly encode a unique and accurate world model.<n>We find that policies trained exclusively on a ttReacher agent's implicit world model are quasi-optimal on out-of-distribution, velocity-based goals despite position-only training.
  arXiv  Detail & Related papers  (2026-06-19T07:26:14Z)
• Generative Criticality in Large Language Model Temperature Scaling [3.472734054649125]
  We propose a statistical-field framework for text generated by large language models (LLMs)<n>We define a susceptibility from the connected two-point correlator and an order parameter from the ensemble-averaged embedding field.<n>We observe a sharp susceptibility peak near a characteristic $T_c$ with power-law-like scaling, a concurrent rapid change in the order parameter, and a collapse onto a single semantic direction below $T_c$.
  arXiv  Detail & Related papers  (2026-06-04T14:43:32Z)
• Universal Spin Squeezing Dynamical Phase Transitions across Lattice Geometries, Dimensions, and Microscopic Couplings [0.0]
  Bogoliubov theory recovers the previously identified scaling $a_Z* propto L$ in the long-range interacting regime $ d+2$.<n>This uncovers a previously unrecognized sub-linear regime for short-range interactions.<n>These findings provide a versatile route toward controlling entanglement generation in Rydberg-array, polar molecule, and trapped-ion platforms.
  arXiv  Detail & Related papers  (2026-05-13T18:00:07Z)
• Attributing Emergence in Million-Agent Systems [68.53670424791751]
  Large language models (LLMs) can simulate human-like reasoning and decision-making in individual agents.<n>Such studies require attributing macro emergence to individual agents.<n>We adapt Aumann--Shapley path-integral attribution to LLM-powered MAS at million-agent scale.
  arXiv  Detail & Related papers  (2026-05-12T01:49:41Z)
• High-Probability Convergence in Decentralized Stochastic Optimization with Gradient Tracking [69.90407799170687]
  We study high-probability (HP) convergence guarantees in decentralized optimization.<n>Our results establish the same conditions on the horizon as in achieving comparable times.
  arXiv  Detail & Related papers  (2026-04-30T22:45:21Z)
• Unsupervised Discovery of Intermediate Phase Order in the Frustrated $J_1$-$J_2$ Heisenberg Model via Prometheus Framework [0.0]
  We apply the Prometheus variational autoencoder framework to explore the $J_1$-$J$ phase diagram.<n>We identify the structure factor $S(,)$ and $S(,)$ as the dominant order parameters.<n>This work establishes a scalable pathway for applying machine learning to frustrated quantum systems.
  arXiv  Detail & Related papers  (2026-02-25T00:44:51Z)
• Graphon Mean-Field Subsampling for Cooperative Heterogeneous Multi-Agent Reinforcement Learning [19.98996237281175]
  We introduce $texttGMFS$, a $textbfG$raphon $textbfM$ean-$textbfF$ield $textbfS$ubsampling framework for scalable cooperative MARL with heterogeneous agent interactions.<n>By subsampling $$ agents according to interaction strength, we approximate the graphon-weighted mean-field and learn a policy with sample complexity.<n>We verify our theory with numerical simulations in robotic coordination, showing that $textttGMFS$ achieves near-optimal performance
  arXiv  Detail & Related papers  (2026-02-18T05:34:07Z)
• Phase Transition for Budgeted Multi-Agent Synergy [41.486076708302456]
  Multi-agent systems can improve reliability, yet under a fixed inference budget they often help, saturate, or even collapse.<n>We develop a minimal and calibratable theory that predicts these regimes from three binding constraints of modern agent stacks.
  arXiv  Detail & Related papers  (2026-01-24T05:32:50Z)
• Generating consensus and dissent on massive discussion platforms with an $O(N)$ semantic-vector model [0.0]
  We introduce a dynamical system to drive the aggregation of semantically similar ideas.<n>Our results show that $> 0$ drives the system toward a ferromagnetic-like phase (global consensus), while $ 0$ induces an antiferromagnetic-like state (maximum dissent)<n>This framework offers a controllable method for managing the tradeoff between cohesion and diversity in Collective Intelligence platforms.
  arXiv  Detail & Related papers  (2026-01-20T13:08:00Z)
• Global Convergence of Gradient EM for Over-Parameterized Gaussian Mixtures [53.51230405648361]
  We study the dynamics of gradient EM and employ tensor decomposition to characterize the geometric landscape of the likelihood loss.<n>This is the first global convergence and recovery result for EM or gradient EM beyond the special case of $m=2$.
  arXiv  Detail & Related papers  (2025-06-06T23:32:38Z)
• Clustered Switchback Designs for Experimentation Under Spatio-temporal Interference [44.644520116360106]
  We estimate the global average treatment effect (GATE), the difference between average outcomes having exposed all units at all times to treatment or to control.<n>We propose a clustered switchback design, where units are grouped into clusters and time steps are grouped into blocks.<n>We show that for graphs that admit good clustering, a truncated Horvitz-Thompson estimator achieves a $tilde O(1/NT)$ mean squared error (MSE)<n>Our results simultaneously generalize the results from citethu2022switchback,ugander2013graph and citetleung2022rate
  arXiv  Detail & Related papers  (2023-12-25T01:00:58Z)
• Ancilla quantum measurements on interacting chains: Sensitivity of entanglement dynamics to the type and concentration of detectors [46.76612530830571]
  We consider a quantum many-body lattice system that is coupled to ancillary degrees of freedom (detectors'') We explore the dynamics of density and of entanglement entropy in the chain, for various values of $rho_a$ and $M$.
  arXiv  Detail & Related papers  (2023-11-21T21:41:11Z)
• Scalable Primal-Dual Actor-Critic Method for Safe Multi-Agent RL with General Utilities [12.104551746465932]
  We investigate safe multi-agent reinforcement learning, where agents seek to collectively maximize an aggregate sum of local objectives while satisfying their own safety constraints. Our algorithm converges to a first-order stationary point (FOSP) at the rate of $mathcalOleft(T-2/3right)$. In the sample-based setting, we demonstrate that, with high probability, our algorithm requires $widetildemathcalOleft(epsilon-3.5right)$ samples to achieve an $epsilon$-FOSP.
  arXiv  Detail & Related papers  (2023-05-27T20:08:35Z)
• Model-Based Multi-Agent RL in Zero-Sum Markov Games with Near-Optimal Sample Complexity [67.02490430380415]
  We show that model-based MARL achieves a sample complexity of $tilde O(|S||B|(gamma)-3epsilon-2)$ for finding the Nash equilibrium (NE) value up to some $epsilon$ error. We also show that such a sample bound is minimax-optimal (up to logarithmic factors) if the algorithm is reward-agnostic, where the algorithm queries state transition samples without reward knowledge.
  arXiv  Detail & Related papers  (2020-07-15T03:25:24Z)

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.