arXiv:2607. 20887v1 Announce Type: cross Abstract: Model merging combines independently trained or fine-tuned models, but pairwise alignability does not imply globally consistent alignment.
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 AI.
arXiv:2605. 18838v3 Announce Type: replace-cross Abstract: Scaling laws predict loss from compute but not how capabilities interact.
arXiv:2609.13874v1 Announce Type: new Abstract: Multimodal clinical AI typically assumes that the waveform, report, metadata, and downstream predictions attached to a record belong to the same patien...
arXiv:2607. 16821v1 Announce Type: cross Abstract: Task arithmetic, sequential fine-tuning, activation steering, and first-order random search all operate through relatively small perturbations around an already trained checkpoint, and they rely on different local approximations: individual perturbations should be first-order predictable, task updates should compose with controlled interference, useful tangent structure should be stable and possible to estimate, and weight edits should have counterparts in representation space.
arXiv:2606. 31963v1 Announce Type: new Abstract: Modern LLM workflows move coordinate-indexed objects across checkpoints: steering vectors, sparse autoencoders, top-$k$ neuron sets, attribution lists, and merge alignments.
The paper investigates how spectral statistics, particularly RankMe and covariance effective rank, behave when large language models (LLMs) undergo post‑training in non‑stationary environments. In controlled experiments on Qwen3‑0.6B, data duplication causes a 75% increase in held‑out loss while simultaneously raising both original and centred RankMe, indicating spectral dispersion rather than rank collapse. The authors demonstrate that a two‑sided, multichannel sequential monitor can detect various damage regimes but does not predict held‑out loss earlier and requires healthy calibration data.
arXiv:2607. 07665v1 Announce Type: new Abstract: Classifier-free guidance (CFG) is the standard way to strengthen class-conditioning in diffusion and flow-matching samplers, yet at large guidance it oversaturates and destabilizes, symptoms practitioners suppress with more steps or limited-interval schedules.