Detecting Mode Collapse in Language Models via Narration

• URL: http://arxiv.org/abs/2402.04477v1
• Date: Tue, 6 Feb 2024 23:52:58 GMT
• Title: Detecting Mode Collapse in Language Models via Narration
• Authors: Sil Hamilton
• Abstract summary: We study 4,374 stories sampled from three OpenAI language models. We show successive versions of GPT-3 suffer from increasing degrees of "mode collapse" Our method and results are significant for researchers seeking to employ language models in sociological simulations.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Few-Shot Detection of Machine-Generated Text using Style Representations [4.326503887981912]
  Language models that convincingly mimic human writing pose a significant risk of abuse. We propose to leverage representations of writing style estimated from human-authored text. We find that features effective at distinguishing among human authors are also effective at distinguishing human from machine authors.
  arXiv  Detail & Related papers  (2024-01-12T17:26:51Z)
• ABINet++: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Spotting [121.11880210592497]
  We argue that the limited capacity of language models comes from 1) implicit language modeling; 2) unidirectional feature representation; and 3) language model with noise input. We propose an autonomous, bidirectional and iterative ABINet++ for scene text spotting.
  arXiv  Detail & Related papers  (2022-11-19T03:50:33Z)
• Cloning Ideology and Style using Deep Learning [0.0]
  Research focuses on text generation based on the ideology and style of a specific author, and text generation on a topic that was not written by the same author in the past. Bi-LSTM model is used to make predictions at the character level, during the training corpus of a specific author is used along with the ground truth corpus. A pre-trained model is used to identify the sentences of ground truth having contradiction with the author's corpus to make our language model inclined.
  arXiv  Detail & Related papers  (2022-10-25T11:37:19Z)
• Model Criticism for Long-Form Text Generation [113.13900836015122]
  We apply a statistical tool, model criticism in latent space, to evaluate the high-level structure of generated text. We perform experiments on three representative aspects of high-level discourse -- coherence, coreference, and topicality. We find that transformer-based language models are able to capture topical structures but have a harder time maintaining structural coherence or modeling coreference.
  arXiv  Detail & Related papers  (2022-10-16T04:35:58Z)
• Integrating Linguistic Theory and Neural Language Models [2.870517198186329]
  I present several case studies to illustrate how theoretical linguistics and neural language models are still relevant to each other. This thesis contributes three studies that explore different aspects of the syntax-semantics interface in language models.
  arXiv  Detail & Related papers  (2022-07-20T04:20:46Z)
• Quark: Controllable Text Generation with Reinforced Unlearning [68.07749519374089]
  Large-scale language models often learn behaviors that are misaligned with user expectations. We introduce Quantized Reward Konditioning (Quark), an algorithm for optimizing a reward function that quantifies an (un)wanted property. For unlearning toxicity, negative sentiment, and repetition, our experiments show that Quark outperforms both strong baselines and state-of-the-art reinforcement learning methods.
  arXiv  Detail & Related papers  (2022-05-26T21:11:51Z)
• On Advances in Text Generation from Images Beyond Captioning: A Case Study in Self-Rationalization [89.94078728495423]
  We show that recent advances in each modality, CLIP image representations and scaling of language models, do not consistently improve multimodal self-rationalization of tasks with multimodal inputs. Our findings call for a backbone modelling approach that can be built on to advance text generation from images and text beyond image captioning.
  arXiv  Detail & Related papers  (2022-05-24T00:52:40Z)
• Testing the Ability of Language Models to Interpret Figurative Language [69.59943454934799]
  Figurative and metaphorical language are commonplace in discourse. It remains an open question to what extent modern language models can interpret nonliteral phrases. We introduce Fig-QA, a Winograd-style nonliteral language understanding task.
  arXiv  Detail & Related papers  (2022-04-26T23:42:22Z)
• Estimating the Personality of White-Box Language Models [0.589889361990138]
  Large-scale language models, which are trained on large corpora of text, are being used in a wide range of applications everywhere. Existing research shows that these models can and do capture human biases. Many of these biases, especially those that could potentially cause harm, are being well-investigated. However, studies that infer and change human personality traits inherited by these models have been scarce or non-existent.
  arXiv  Detail & Related papers  (2022-04-25T23:53:53Z)
• Uncovering Constraint-Based Behavior in Neural Models via Targeted Fine-Tuning [9.391375268580806]
  We show that competing linguistic processes within a language obscure underlying linguistic knowledge. While human behavior has been found to be similar across languages, we find cross-linguistic variation in model behavior. Our results suggest that models need to learn both the linguistic constraints in a language and their relative ranking, with mismatches in either producing non-human-like behavior.
  arXiv  Detail & Related papers  (2021-06-02T14:52:11Z)
• Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition [80.446770909975]
  Linguistic knowledge is of great benefit to scene text recognition. How to effectively model linguistic rules in end-to-end deep networks remains a research challenge. We propose an autonomous, bidirectional and iterative ABINet for scene text recognition.
  arXiv  Detail & Related papers  (2021-03-11T06:47:45Z)

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.