arXiv:2607. 09487v1 Announce Type: new Abstract: Within-class variance in language-model representations is commonly read as incomplete neural collapse.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.
The paper investigates why large language models (LLMs) fail to maintain continuous mixtures of token embeddings—used in latent-state reasoning—to preserve multiple reasoning paths. Through theory and experiments, it identifies three failure sources: transformer geometry distortion, amplification or contraction dynamics from softmax and autoregressive feedback, and the need for context-dependent corrections that scale with mixture size. Empirical results confirm the predicted transition between contraction and amplification and show pretrained models largely fall on the amplifying side.
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.
arXiv:2609.36455v1 Announce Type: new Abstract: Large language models are thought to represent features by vectors in a hidden space of dimension given by the model's width. Superposition, in which m...
arXiv:2609. 31181v1 Announce Type: new Abstract: Black-box model identification works by scoring a model's response to natural-language prompts.
arXiv:2606. 07559v1 Announce Type: cross Abstract: Fine-tuning a language model on contexts whose correct completion has a near-synonym competitor often fails silently.
arXiv:2606. 07559v2 Announce Type: replace-cross Abstract: Fine-tuning a language model often fails silently when its correct completion must outrank a near-synonym competitor.