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.
Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.
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:2608. 10251v1 Announce Type: cross Abstract: A transformer's answer lives on one axis: the direction its unembedding reads.
arXiv:2608. 05164v1 Announce Type: cross Abstract: Independently trained large language models may develop shared internal representations of semantic concepts despite architectural differences -- but whether this geometric similarity has functional consequences for cross-model behavioural control remains untested.
arXiv:2602. 22600v2 Announce Type: replace-cross Abstract: Training selects for behavior, not circuitry: many weight configurations can implement the same function.
Neural Collapse predicts that balanced one-hot classification pushes model representations to be equally far from each other; a symmetric configuration that depends only on the output label and ignores any semantic similarity in the inputs. This creates a puzzle: next-token prediction language models are trained predominantly (as context length increases) with one-hot labels: the same context is very unlikely to appear twice in training with different labels.
arXiv:2606. 26749v1 Announce Type: new Abstract: Neural Collapse predicts that balanced one-hot classification pushes model representations to be equally far from each other; a symmetric configuration that depends only on the output label and ignores any semantic similarity in the inputs.