arXiv:2609.11149v3 Announce Type: replace-cross Abstract: How fast does a language model degrade when trained on its own outputs? Theory traces it to gradually accumulating errors, while experiments...
The paper investigates neural text degeneration by measuring the fixed‑point structure of short‑window argmax maps across 17 pretrained models, using 96 random two‑token starts without prompts. It finds a stable four‑way classification that varies across model families and scales, with some models funneling to a single endpoint token while others do not, and shows that this behavior is not solely determined by training data or corpus frequency. The study demonstrates that repetition phenomena are not uniformly explained by either training data or network architecture alone, highlighting the complexity of neural text generation dynamics.
The paper investigates how recursive contamination—retraining language models on their own generated text—affects output diversity across 13 publicly released checkpoints. Using a fixed contamination protocol over five generations, the authors find a wide spread in 4‑gram diversity (0.187 to 0.940), indicating that some models collapse into repetitive fragments while others remain largely unaffected. The study shows that a model’s susceptibility to collapse is an intrinsic property of the checkpoint, not predicted by parameter scale or static indicators, and that simple interventions such as tightening top‑p sampling can significantly slow or halt collapse.
The paper investigates whether an oligopolistic concentration of generative AI models accelerates or steers the phenomenon of model collapse when models are recursively trained on each other’s outputs. Using controlled ecosystems of 13 open‑source models and an injected probe that pushes one model’s market share to 90%, the authors find that varying market concentration has little effect on the speed or final state of collapse. Instead, the pace of collapse is largely determined by which models supply the training pool and how susceptible those models are to being carried along, with human‑written text in the pool roughly halving the drift.
arXiv:2506.17871v4 Announce Type: replace-cross Abstract: Despite their impressive capabilities, aligned large language models (LLMs) often generate outputs that lack diversity. What drives this cons...
arXiv:2609.08475v1 Announce Type: cross Abstract: Large language models have collapsed the cost of producing lexically elaborate prose, and whether peer reviewers still reward it is a question about...
arXiv:2608. 11694v1 Announce Type: cross Abstract: A benchmark score comes from a single phrasing of each problem.
Appending a two-word confirmation tag to a decision question -- "Is X the better choice? " versus "X is the better choice, right?
arXiv:2606. 01202v1 Announce Type: new Abstract: Language models do not simply choose an answer at the output layer.
arXiv:2609.39702v1 Announce Type: new Abstract: Adapting a language model to a specialized corpus means choosing which instruction-tuning tasks to train on under a fixed budget, and testing one choic...
arXiv:2607. 23976v1 Announce Type: cross Abstract: Appending a two-word confirmation tag to a decision question -- "Is X the better choice?
arXiv:2608. 04021v1 Announce Type: cross Abstract: Cloze-style probes that vary how often a target token appears implicitly assume that more copies of a target affect prediction the same way regardless of where the readout slot sits.