Theoretical Proof that Generated Text in the Corpus Leads to the Collapse of Auto-regressive Language Models

• URL: http://arxiv.org/abs/2412.14872v2
• Date: Tue, 11 Feb 2025 12:25:11 GMT
• Title: Theoretical Proof that Generated Text in the Corpus Leads to the Collapse of Auto-regressive Language Models
• Authors: Lecheng Wang, Xianjie Shi, Ge Li, Jia Li, Xuanming Zhang, Yihong Dong, Wenpin Jiao, Hong Mei,
• Abstract summary: This paper presents theoretical proof that once a corpus (such as the World Wide Web) begins to incorporate generated text, LM collapse is bound to occur.<n>We express our concerns about the current situation in which an increasing amount of generated text may be used in LM training.
• Score: 26.117724170912552
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Auto-regressive language models (LMs) have been widely used to generate text on the World Wide Web. The generated text is often collected into the training corpus of the next generations of LMs. Previous work experimentally found that LMs collapse when trained on recursively generated text. This paper presents theoretical proof that once a corpus (such as the World Wide Web) begins to incorporate generated text, and the training text of each LM is sampled from this corpus, then no matter how small the amount of text generated by each LM that enters the corpus is, after a sufficient amount of time, LM collapse is bound to occur. Our proof is validated by a series of experiments showing that the collapsed LMs perform no better than an untrained LM with randomly initialized parameters. By proving the existence of LM collapse, we express our concerns about the current situation in which an increasing amount of generated text may be used in LM training. The source code is available in the online data warehouse: https://github.com/wanglc02/generated-data

Related papers

• PuckTrick: A Library for Making Synthetic Data More Realistic [46.198289193451146]
  We introduce Pucktrick, a Python library designed to systematically contaminate synthetic datasets by introducing controlled errors.<n>We evaluate the impact of systematic data contamination on model performance.<n>Our findings demonstrate that ML models trained on contaminated synthetic data outperform those trained on purely synthetic, error-free data.
  arXiv  Detail & Related papers  (2025-06-23T10:51:45Z)
• Hey, That's My Data! Label-Only Dataset Inference in Large Language Models [63.35066172530291]
  CatShift is a label-only dataset-inference framework.<n>It capitalizes on catastrophic forgetting: the tendency of an LLM to overwrite previously learned knowledge when exposed to new data.
  arXiv  Detail & Related papers  (2025-06-06T13:02:59Z)
• A Theoretical Perspective: How to Prevent Model Collapse in Self-consuming Training Loops [55.07063067759609]
  High-quality data is essential for training large generative models, yet the vast reservoir of real data available online has become nearly depleted.<n>Models increasingly generate their own data for further training, forming Self-consuming Training Loops (STLs)<n>Some models degrade or even collapse, while others successfully avoid these failures, leaving a significant gap in theoretical understanding.
  arXiv  Detail & Related papers  (2025-02-26T06:18:13Z)
• Machine-generated text detection prevents language model collapse [17.34282527020344]
  We investigate the impact of decoding strategy on model collapse.<n>We train a machine-generated text detector and propose an importance sampling approach to alleviate model collapse.
  arXiv  Detail & Related papers  (2025-02-21T18:22:36Z)
• Understanding Synthetic Context Extension via Retrieval Heads [51.8869530817334]
  We investigate fine-tuning on synthetic data for three long-context tasks that require retrieval and reasoning.<n>We find that models trained on synthetic data fall short of the real data, but surprisingly, the mismatch can be interpreted.<n>Our results shed light on how to interpret synthetic data fine-tuning performance and how to approach creating better data for learning real-world capabilities over long contexts.
  arXiv  Detail & Related papers  (2024-10-29T17:55:00Z)
• Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating World [19.266191284270793]
  generative machine learning models are pretrained on web-scale datasets containing data generated by earlier models.<n>Some prior work warns of "model collapse" as the web is overwhelmed by synthetic data.<n>We report experiments on three ways of using data (training-workflows) across three generative model task-settings.
  arXiv  Detail & Related papers  (2024-10-22T05:49:24Z)
• Entropy Law: The Story Behind Data Compression and LLM Performance [115.70395740286422]
  We find that model performance is negatively correlated to the compression ratio of training data, which usually yields a lower training loss. Based on the findings of the entropy law, we propose a quite efficient and universal data selection method. We also present an interesting application of entropy law that can detect potential performance risks at the beginning of model training.
  arXiv  Detail & Related papers  (2024-07-09T08:14:29Z)
• Evaluating $n$-Gram Novelty of Language Models Using Rusty-DAWG [57.14250086701313]
  We investigate the extent to which modern LMs generate $n$-grams from their training data. We develop Rusty-DAWG, a novel search tool inspired by indexing of genomic data.
  arXiv  Detail & Related papers  (2024-06-18T21:31:19Z)
• Beyond Model Collapse: Scaling Up with Synthesized Data Requires Verification [11.6055501181235]
  We investigate the use of verification on synthesized data to prevent model collapse. We show that verifiers, even imperfect ones, can indeed be harnessed to prevent model collapse.
  arXiv  Detail & Related papers  (2024-06-11T17:46:16Z)
• Nearest Neighbor Speculative Decoding for LLM Generation and Attribution [87.3259169631789]
  Nearest Speculative Decoding (NEST) is capable of incorporating real-world text spans of arbitrary length into the LM generations and providing attribution to their sources. NEST significantly enhances the generation quality and attribution rate of the base LM across a variety of knowledge-intensive tasks. In addition, NEST substantially improves the generation speed, achieving a 1.8x speedup in inference time when applied to Llama-2-Chat 70B.
  arXiv  Detail & Related papers  (2024-05-29T17:55:03Z)
• Reliable, Adaptable, and Attributable Language Models with Retrieval [144.26890121729514]
  Parametric language models (LMs) are trained on vast amounts of web data. They face practical challenges such as hallucinations, difficulty in adapting to new data distributions, and a lack of verifiability. We advocate for retrieval-augmented LMs to replace parametric LMs as the next generation of LMs.
  arXiv  Detail & Related papers  (2024-03-05T18:22:33Z)
• Deductive Closure Training of Language Models for Coherence, Accuracy, and Updatability [58.582216812183496]
  Language models (LMs) can sometimes generate factually correct text and estimate truth values of individual claims. Current LMs generate incorrect or nonsensical content, and are difficult to edit and bring up to date. We present a method called Deductive Closure Training (DCT) that uses LMs themselves to identify implications of (and contradictions within) the text that they generate.
  arXiv  Detail & Related papers  (2024-01-16T18:58:37Z)
• Retrieval-Pretrained Transformer: Long-range Language Modeling with Self-retrieval [51.437420003471615]
  We propose the Retrieval-Pretrained Transformer (RPT), an architecture and training procedure for jointly training a retrieval-augmented LM from scratch. RPT improves retrieval quality and subsequently perplexity across the board compared to strong baselines.
  arXiv  Detail & Related papers  (2023-06-23T10:18:02Z)
• LeTI: Learning to Generate from Textual Interactions [60.425769582343506]
  We explore LMs' potential to learn from textual interactions (LETI) that not only check their correctness with binary labels but also pinpoint and explain errors in their outputs through textual feedback. Our focus is the code generation task, where the model produces code based on natural language instructions. LETI iteratively fine-tunes the model, using the objective LM, on a concatenation of natural language instructions, LM-generated programs, and textual feedback.
  arXiv  Detail & Related papers  (2023-05-17T15:53:31Z)
• Synthetic data, real errors: how (not) to publish and use synthetic data [86.65594304109567]
  We show how the generative process affects the downstream ML task. We introduce Deep Generative Ensemble (DGE) to approximate the posterior distribution over the generative process model parameters.
  arXiv  Detail & Related papers  (2023-05-16T07:30:29Z)
• Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP [77.817293104436]
  We propose a framework that relies on passing natural language texts in sophisticated pipelines between an LM and an RM. We have written novel DSP programs for answering questions in open-domain, multi-hop, and conversational settings.
  arXiv  Detail & Related papers  (2022-12-28T18:52:44Z)
• Factuality Enhanced Language Models for Open-Ended Text Generation [60.27166549575472]
  We design the FactualityPrompts test set and metrics to measure the factuality of LM generations. We find that larger LMs are more factual than smaller ones, although a previous study suggests that larger LMs can be less truthful in terms of misconceptions. We propose a factuality-enhanced training method that uses TopicPrefix for better awareness of facts and sentence completion.
  arXiv  Detail & Related papers  (2022-06-09T17:16:43Z)
• Neuro-Symbolic Language Modeling with Automaton-augmented Retrieval [129.25914272977542]
  RetoMaton is a weighted finite automaton built on top of the datastore. Traversing this automaton at inference time, in parallel to the LM inference, reduces its perplexity.
  arXiv  Detail & Related papers  (2022-01-28T21:38:56Z)

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.