# Beyond the Hivemind: Escaping LLM Homogeneity via Meta-Persona Anchoring and Sequential Temperature Scaling

arXiv: https://arxiv.org/abs/2608.02618 · HTML: https://mlsanity.com/paper/2608.02618/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 16 references that are in the map.
- First posted on arXiv 31 May 2026.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputs](https://mlsanity.com/paper/2407.01082/) (2024)
- [Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond)](https://mlsanity.com/paper/2510.22954/) (2025)
- [Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity](https://mlsanity.com/paper/2510.01171/) (2025)
- [Towards Understanding the Interplay of Generative Artificial Intelligence and the Internet](https://mlsanity.com/paper/2306.06130/) (2023)
- [Min-k Sampling: Decoupling Truncation from Temperature Scaling via Relative Logit Dynamics](https://mlsanity.com/paper/2604.11012/) (2026)
- [Latent Semantic Manifolds in Large Language Models](https://mlsanity.com/paper/2603.22301/) (2026)
- [Not All Layers Need Tuning: Selective Layer Restoration Recovers Diversity](https://mlsanity.com/paper/2602.06665/) (2026)

## Abstract (arXiv)

Recent studies have identified an ``Artificial Hivemind'' effect in Large Language Models (LLMs) causing models to converge on a narrow, homogenized consensus even for open questions. This semantic collapse limits the diversity of AI, resulting in high inter-response similarity (≈ 0.80-0.90) even under high-temperature sampling. In this paper, we propose a novel mitigation framework to increase diversity: Meta-Persona Anchoring combined with Filtered Temperature Scaling (FTS). Our approach utilizes a two-stage generation process: first, the model is prompted to self-select a unique, idiosyncratic persona to anchor its starting point; second, we apply a dual-stage sampling sieve, utilizing Top-p filtering to preserve grammatical validity followed by extreme temperature scaling (T ≥ 4.0) on the surviving candidates to explore the broadened probability distribution. We evaluate our method using the INFINITY-CHAT dataset on state-of-the-art open weight models under ~20B parameters. Our results demonstrate a significant reduction in semantic convergence, with average pairwise cosine similarity dropping from (≈ 0.85) to (≈ 0.65). Our scheme achieves a majority of questions below the 0.7 threshold, effectively reducing the gap between artificial mode collapse and human-level typological diversity. We provide our implementation as an open-source framework to enable more diverse and creative AI deployments.