Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Shared Nature, Unique Nurture: PRISM for Pluralistic Reasoning via In-context Structure Modeling

Guancheng Tu, Shiyang Zhang, Tianyu Zhang, Yi Zhang, Diji Yang

TL;DR

This work establishes a new paradigm for Pluralistic AI, moving beyond monolithic consensus toward a diverse ecosystem of unique cognitive individuals capable of collective, multi-perspective discovery.

Abstract

Large Language Models (LLMs) are converging towards a singular Artificial Hivemind, where shared Nature (pre-training priors) result in a profound collapse of distributional diversity, limiting the distinct perspectives necessary for creative exploration and scientific discovery. To address this, we propose to equip models with inference-time Nurture (individualized epistemic trajectories) using Epistemic Evolution paradigm, progressing through explore, internalize, and express. We instantiate this via PRISM (Pluralistic Reasoning via In-context Structure Modeling), a model-agnostic system that augments LLM with dynamic On-the-fly Epistemic Graphs. On three creativity benchmarks, PRISM achieves state-of-the-art novelty and significantly expands distributional diversity. Moreover, we evaluate the real-world utility via a challenging rare-disease diagnosis benchmark. Results demonstrate that PRISM successfully uncovers correct long-tail diagnoses that standard LLM miss, confirming that its divergence stems from meaningful exploration rather than incoherent noise. Overall, this work establishes a new paradigm for Pluralistic AI, moving beyond monolithic consensus toward a diverse ecosystem of unique cognitive individuals capable of collective, multi-perspective discovery.

Shared Nature, Unique Nurture: PRISM for Pluralistic Reasoning via In-context Structure Modeling

TL;DR

This work establishes a new paradigm for Pluralistic AI, moving beyond monolithic consensus toward a diverse ecosystem of unique cognitive individuals capable of collective, multi-perspective discovery.

Abstract

Large Language Models (LLMs) are converging towards a singular Artificial Hivemind, where shared Nature (pre-training priors) result in a profound collapse of distributional diversity, limiting the distinct perspectives necessary for creative exploration and scientific discovery. To address this, we propose to equip models with inference-time Nurture (individualized epistemic trajectories) using Epistemic Evolution paradigm, progressing through explore, internalize, and express. We instantiate this via PRISM (Pluralistic Reasoning via In-context Structure Modeling), a model-agnostic system that augments LLM with dynamic On-the-fly Epistemic Graphs. On three creativity benchmarks, PRISM achieves state-of-the-art novelty and significantly expands distributional diversity. Moreover, we evaluate the real-world utility via a challenging rare-disease diagnosis benchmark. Results demonstrate that PRISM successfully uncovers correct long-tail diagnoses that standard LLM miss, confirming that its divergence stems from meaningful exploration rather than incoherent noise. Overall, this work establishes a new paradigm for Pluralistic AI, moving beyond monolithic consensus toward a diverse ecosystem of unique cognitive individuals capable of collective, multi-perspective discovery.
Paper Structure (61 sections, 2 equations, 8 figures, 7 tables)

This paper contains 61 sections, 2 equations, 8 figures, 7 tables.

Table of Contents

  1. Introduction
  2. PRISM for Pluralistic LLMs
  3. Phase I: Cognitive Explosion (Breaking Nature)
  4. Phase II: Epistemic Graph Construction (Structuring Nurture)
  5. Design Rationale: Why Graph Topology?
  6. Phase III: Conditional Generation
  7. Experiments: Creativity and Discovery
  8. Experimental Setup
  9. Distributional Diversity: Artificial Hivemind
  10. Open-Ended Creativity: NoveltyBench
  11. Scientific Discovery: IdeaBench
  12. Real-world Application: Rare Disease Diagnosis
  13. The Dilemma of Conservative Models
  14. Methodology: Disentangling Search, Structure, and Intent
  15. Results: Breaking the Conservative Trap
  16. ...and 46 more sections

Figures (8)

  • Figure 1: Divergent cognitive responses to a common stimulus: Newton's gravity vs. Kolb's candy apple, both inspired by an apple. Analogous to an optical prism, our PRISM framework aims to restore this pluralism by refracting shared pre-training knowledge into distinct, individualized epistemic trajectories.
  • Figure 2: The PRISM Pipeline. The system first performs Cognitive Explosion via wild search to break local relevance, followed by Epistemic Structuring where Context Nodes and Spark Nodes are bridged via cognitive operators (Mapping, Blending, Inversion).
  • Figure 3: PCA visualization of response distributions. PRISM (in dot) produces multi-centered, elongated distributions compared to the concentrated clusters of vanilla generation.
  • Figure 4: Intra-model diversity heatmap on Artificial Hivemind.
  • Figure 5: Inter-model similarity heatmap on Artificial Hivemind.
  • ...and 3 more figures