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...
By Yangze Liu, Zhongyi Han
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.
By Nicol\'as Vera Z\'u\~niga
arXiv:2606. 26050v1 Announce Type: new Abstract: Midway through an ordinary pretraining run, a small language model learns the pronoun-gender rule: cued with a girl's name ("Sue cried because"), it resolves the next pronoun to she, generalizing to held-out probes (0.
By Juliana Li, Diya Sreedhar
arXiv:2603. 11784v2 Announce Type: replace Abstract: As scaling laws push the training of frontier large language models (LLMs) toward ever-growing data requirements, training pipelines are approaching a regime where much of the publicly available online text may be consumed.
By Giorgio Racca, Michal Valko, Amartya Sanyal
Midway through an ordinary pretraining run, a small language model learns the pronoun-gender rule: cued with a girl's name ("Sue cried because"), it resolves the next pronoun to she, generalizing to held-out probes (0. 94 by step 925).
The paper investigates when a language model actually learns task information versus when its readout (unembedding matrix) learns to express that information. Using a new parameter‑trajectory cross‑coding method, the authors find that hidden states can encode task data early in training, but the readout reorganizes only in a narrow window (0.4–1.1% of Pythia’s pretraining). By varying the readout learning rate, they show that faster readout learning reduces the gap between hidden‑state encoding and output expression without affecting final performance.
By Matteo He, William F. Shen, Alex Iacob, Andrej Jovanovic, Xinchi Qiu, Nicholas D. Lane
The paper investigates how preference optimization can inadvertently create a shared directional shift—termed a common‑mode offset—in a model’s hidden states. This offset, which affects both chosen and rejected responses, accumulates without being penalized by the ranking loss, eventually projecting onto the output layer to form a vocabulary well that dominates decoding and leads to generation collapse. Experiments across multiple base‑model lineages and fine‑tuning regimes confirm that masking this offset or using per‑token ranking losses can prevent the collapse, while standard ranking metrics fail to detect the risk.
By Yun Li, Ehsan Javanmardi, Simon Thompson, Jiachen Gong, Qunli Zhang, Zifan Zeng, Shiming Liu, Peng Wang, Zixuan Guo, Manabu Tsukada
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...
By Chenghao Yang, Sida Li, Ari Holtzman
The study investigates whether the phenomenon of capitulation—where language models abandon correct answers under user pushback—can be identified as a linear direction in the residual stream and whether this direction emerges with model scale. Analyzing eight instruction‑tuned models from 0.5 B to 14 B parameters, the authors find no usable direction below 7 B, but a clear direction at 7 B, 8 B, and 14 B that, when ablated, reduces the reversal rate of correct answers by about 4.8 percentage points. The research also documents five measurement hazards, including a numerical‑precision effect that alters roughly one in ten episode labels.
By Saad Aamir, Muhammad Awais Bin Adil
The paper investigates how generative models that learn from the outputs of earlier models lose rare knowledge over successive generations, a phenomenon termed model collapse. It argues that the problem lies in the supply of fresh data, which is often fixed and external, and proposes that making this supply endogenous—allowing users to query models and learn from them—can mitigate collapse. Experiments confirm that when the world changes, copying lineages require more new data, while lineages that restart from fixed pretrained weights remain stable, and that user checks and relearning are essential to avoid the paradox.
By Nam H. Le
Sycophancy in language models is well documented as a behaviour but poorly localised as a mechanism. We ask whether capitulation under user pushback, the abandonment of a correct answer when a user ob...
arXiv:2607. 25063v1 Announce Type: new Abstract: Developers judge a model checkpoint by how it behaves.
By Cen Lu, Yung-Chen Tang, Andrea Cavallaro