# Detecting Mode Collapse in Language Models via Narration

arXiv: https://arxiv.org/abs/2402.04477 · HTML: https://mlsanity.com/paper/2402.04477/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 32 references that are in the map.
- First posted on arXiv 6 Feb 2024.

## Builds on (most cited first, 9 of 9)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [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)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Large Language Models Can Be Easily Distracted by Irrelevant Context](https://mlsanity.com/paper/2302.00093/) (2023)
- [Out of One, Many: Using Language Models to Simulate Human Samples](https://mlsanity.com/paper/2209.06899/) (2022)
- [Social Simulacra: Creating Populated Prototypes for Social Computing Systems](https://mlsanity.com/paper/2208.04024/) (2022)
- [Supervising strong learners by amplifying weak experts](https://mlsanity.com/paper/1810.08575/) (2018)

## Built on by (most cited first, 24 of 24)

- [NoveltyBench: Evaluating Language Models for Humanlike Diversity](https://mlsanity.com/paper/2504.05228/) (2025)
- [WARP: On the Benefits of Weight Averaged Rewarded Policies](https://mlsanity.com/paper/2406.16768/) (2024)
- [Bias Amplification: Large Language Models as Increasingly Biased Media](https://mlsanity.com/paper/2410.15234/) (2024)
- [FactorSim: Generative Simulation via Factorized Representation](https://mlsanity.com/paper/2409.17652/) (2024)
- [Diversity-Rewarded CFG Distillation](https://mlsanity.com/paper/2410.06084/) (2024)
- [Optimizing Diversity and Quality through Base-Aligned Model Collaboration](https://mlsanity.com/paper/2511.05650/) (2025)
- [Leveraging Large Language Models to Detect npm Malicious Packages](https://mlsanity.com/paper/2403.12196/) (2024)
- [Domain Regeneration: How well do LLMs match syntactic properties of text domains?](https://mlsanity.com/paper/2505.07784/) (2025)
- [CORE: Measuring Multi-Agent LLM Interaction Quality under Game-Theoretic Pressures](https://mlsanity.com/paper/2508.11915/) (2025)
- [Narrative Theory-Driven LLM Methods for Automatic Story Generation and Understanding: A Survey](https://mlsanity.com/paper/2602.15851/) (2026)
- [Annotations Mitigate Post-Training Mode Collapse](https://mlsanity.com/paper/2605.09995/) (2026)
- [The Homogenization Problem in LLMs: Towards Meaningful Diversity in AI Safety](https://mlsanity.com/paper/2601.06116/) (2026)
- [Structuring the Space of Sociotechnical Alignment](https://mlsanity.com/paper/2607.01250/) (2026)
- [The Challenge and Reward of Fair Play in Narrative: A Computational Approach](https://mlsanity.com/paper/2507.13841/) (2025)
- [Reducing the Probability of Undesirable Outputs in Language Models Using Probabilistic Inference](https://mlsanity.com/paper/2510.21184/) (2025)
- [STEER: Inference-Time Risk Control via Constrained Quality-Diversity Search](https://mlsanity.com/paper/2602.02862/) (2026)
- [Beyond One Output: Visualizing and Comparing Distributions of Language Model Generations](https://mlsanity.com/paper/2604.18724/) (2026)
- [Elias in the Lighthouse, Again? Diagnosing Low Diversity in LLM Stories](https://mlsanity.com/paper/2605.26492/) (2026)
- [GENIE: A Fine-Grained Measure for Novelty](https://mlsanity.com/paper/2606.12790/) (2026)
- [Collective cooperation without individual fidelity in LLM agents](https://mlsanity.com/paper/2606.30454/) (2026)
- [Can We Trust LLM's Logic? Quantifying Uncertainty, Coherence, and Robustness via a Graph-Based Framework](https://mlsanity.com/paper/2607.08017/) (2026)
- [ARC: Fair Relative Advantage Comparison in Open-Ended Real-World Interaction](https://mlsanity.com/paper/2608.13622/) (2026)
- [Narrowing the Horizon: Quantifying Topic Saliency Shifts in Generative Monoculture](https://mlsanity.com/paper/2609.35302/) (2026)
- [Beyond Mode Collapse: Generating Diverse Synthetic Expert Conversations via Generative Flow Networks](https://mlsanity.com/paper/2609.38359/) (2026)

## Abstract (arXiv)

No two authors write alike. Personal flourishes invoked in written narratives, from lexicon to rhetorical devices, imply a particular author--what literary theorists label the implied or virtual author; distinct from the real author or narrator of a text. Early large language models trained on unfiltered training sets drawn from a variety of discordant sources yielded incoherent personalities, problematic for conversational tasks but proving useful for sampling literature from multiple perspectives. Successes in alignment research in recent years have allowed researchers to impose subjectively consistent personae on language models via instruction tuning and reinforcement learning from human feedback (RLHF), but whether aligned models retain the ability to model an arbitrary virtual author has received little scrutiny. By studying 4,374 stories sampled from three OpenAI language models, we show successive versions of GPT-3 suffer from increasing degrees of "mode collapse" whereby overfitting the model during alignment constrains it from generalizing over authorship: models suffering from mode collapse become unable to assume a multiplicity of perspectives. Our method and results are significant for researchers seeking to employ language models in sociological simulations.