Metraipolis › archive › dc_5fdf7ddb2595

Spectral Smoothing and Spiking Perturbations Reconfigure Capability Profiles and Selective Task Performance in Large Language Models — arXiv:2610.30676

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30676
License (as archived)
CC0-1.0
Archived
2026-10-23T15:15:03.438Z
SHA-256 of the archived text
2c540a66895835f7b1dd659025d724dda4f2186b6b129cef2f3c82485dcef149
Stable URL of this page
https://metraipolis.com/documento/dc_5fdf7ddb2595

How to cite

Spectral Smoothing and Spiking Perturbations Reconfigure Capability Profiles and Selective Task Performance in Large Language Models — arXiv:2610.30676 — arXiv (en), https://arxiv.org/abs/2610.30676 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5fdf7ddb2595 (Metraipolis archive dc_5fdf7ddb2595, sha256 2c540a66895835f7b1dd659025d724dda4f2186b6b129cef2f3c82485dcef149)
BibTeX
@misc{dc5fdf7ddb2595,
  title = {Spectral Smoothing and Spiking Perturbations Reconfigure Capability Profiles and Selective Task Performance in Large Language Models — arXiv:2610.30676},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30676},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5fdf7ddb2595; sha256 2c540a66895835f7b1dd659025d724dda4f2186b6b129cef2f3c82485dcef149},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30676. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Singular-value spectra encode important structure in transformer feed-forward projections, yet their causal role in downstream behavior remains underexplored. We study post-hoc spectral smoothing and spiking interventions on the up_proj and down_proj matrices. Positive perturbation stretches deviations from the geometric mean, making large singular values larger and small ones smaller (spiking), and vice versa. The intervention preserves singular vectors and the geometric mean of the singular values. Across five open Llama and Qwen instruction models, we evaluate the perturbed systems on capability-labeled subtasks, standard and chain-of-thought reasoning benchmarks, judge-scored truthfulness, hallucination, and safety tasks, and TRAIT personality probes. To interpret score shifts, we fit responses in a three-dimensional capability basis consisting of factual knowledge, language understanding, and deductive reasoning. The induced changes suggest structured rather than purely random sensitivity: large positive perturbations more reliably suppress language and deductive responses, moderate negative perturbations can sometimes improve factual or task-local performance on the evaluated grids, and sensitivity varies strongly by model family, benchmark module, and underlying layerwise spectral inequality. These results position spectral perturbation as a lightweight intervention for probing how capabilities are organized and traded off in large language models.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.30676 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-23T15:09:08.543Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).