# CreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse

arXiv: https://arxiv.org/abs/2607.01433 · HTML: https://mlsanity.com/paper/2607.01433/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 23 of its 51 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 23 of 23)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [A Simple and Effective Pruning Approach for Large Language Models](https://mlsanity.com/paper/2306.11695/) (2023)
- [Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers](https://mlsanity.com/paper/2409.04109/) (2024)
- [Programming Refusal with Conditional Activation Steering](https://mlsanity.com/paper/2409.05907/) (2024)
- [SciMON: Scientific Inspiration Machines Optimized for Novelty](https://mlsanity.com/paper/2305.14259/) (2023)
- [Open Problems in Mechanistic Interpretability](https://mlsanity.com/paper/2501.16496/) (2025)
- [Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond)](https://mlsanity.com/paper/2510.22954/) (2025)
- [Is Temperature the Creativity Parameter of Large Language Models?](https://mlsanity.com/paper/2405.00492/) (2024)
- [The Ideation-Execution Gap: Execution Outcomes of LLM-Generated versus Human Research Ideas](https://mlsanity.com/paper/2506.20803/) (2025)
- [Questioning Representational Optimism in Deep Learning: The Fractured Entangled Representation Hypothesis](https://mlsanity.com/paper/2505.11581/) (2025)
- [Steering Language Models with Weight Arithmetic](https://mlsanity.com/paper/2511.05408/) (2025)
- [Steering Vector Fields for Context-Aware Inference-Time Control in Large Language Models](https://mlsanity.com/paper/2602.01654/) (2026)
- [Spark: A System for Scientifically Creative Idea Generation](https://mlsanity.com/paper/2504.20090/) (2025)
- [Math Neurosurgery: Isolating Language Models' Math Reasoning Abilities Using Only Forward Passes](https://mlsanity.com/paper/2410.16930/) (2024)
- [What Can We Actually Steer? A Multi-Behavior Study of Activation Control](https://mlsanity.com/paper/2511.18284/) (2025)
- [Transformational Creativity in Science: A Graphical Theory](https://mlsanity.com/paper/2504.18687/) (2025)
- [Annotations Mitigate Post-Training Mode Collapse](https://mlsanity.com/paper/2605.09995/) (2026)
- [Steering Large Language Models to Evaluate and Amplify Creativity](https://mlsanity.com/paper/2412.06060/) (2024)
- [Enhancing Diversity of LLM-Generated Educational Tasks](https://mlsanity.com/paper/2512.23601/) (2025)

## Built on by (most cited first, 2 of 2)

- [Breaking Homogeneity: Diversifying Persona Sets for Creative LLM Outputs](https://mlsanity.com/paper/2609.30492/) (2026)
- [PainterBench: A Figural Divergent-Thinking Benchmark for Tool-Using Language Models](https://mlsanity.com/paper/2609.34195/) (2026)

## Abstract (arXiv)

Divergent thinking is a crucial aspect of creativity, yet large language models (LLMs) tend to consistently generate similar responses to open-ended questions, in what has been termed the artificial hivemind effect. Here, we introduce CreativityNeuro, a data-free method for enhancing divergent thinking in LLMs via contrastive weight steering. We evaluate our method across multiple creativity assessments and report several main findings. On the Divergent Association Task (DAT), a vocabulary-space creativity test, CreativityNeuro improves performance by up to 14 human percentile points. Next, in a large-scale human evaluation (N=720) on the Alternative Uses Test (AUT) and the Task Task, CreativityNeuro achieves significant improvements in originality, surprise, and creativity, transferring to longer-form and more open-ended tasks. Importantly, we find that across all three tasks, CreativityNeuro demonstrably reduces measures of mode collapse. Moreover, activation steering achieves comparable performance to CreativityNeuro on the DAT, but it does not transfer to the AUT and Task Task, demonstrating the effectiveness of weight-space steering in generalizing to unseen tasks. In conclusion, CreativityNeuro improves divergent thinking and reduces mode collapse without requiring behavioral data, re-training, or gradient-based fine-tuning, providing a straightforward way to enhance LLM performance in creative domains.