Metraipolis › archive › dc_2bb080bc1bf1

ForeSight: Enhancing Risk Monitoring via Early Safety Signal Distillation — arXiv:2609.13737

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.13737
License (as archived): CC0-1.0
Archived: 2026-10-15T23:14:38.178Z
SHA-256 of the archived text: df823aad4dd768fa9862eb0a1e01131f976698b029a0688445e6f459b5a881be
Stable URL of this page: https://metraipolis.com/documento/dc_2bb080bc1bf1

How to cite

ForeSight: Enhancing Risk Monitoring via Early Safety Signal Distillation — arXiv:2609.13737 — arXiv (en), https://arxiv.org/abs/2609.13737 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2bb080bc1bf1 (Metraipolis archive dc_2bb080bc1bf1, sha256 df823aad4dd768fa9862eb0a1e01131f976698b029a0688445e6f459b5a881be)

BibTeX

@misc{dc2bb080bc1bf1,
  title = {ForeSight: Enhancing Risk Monitoring via Early Safety Signal Distillation — arXiv:2609.13737},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13737},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2bb080bc1bf1; sha256 df823aad4dd768fa9862eb0a1e01131f976698b029a0688445e6f459b5a881be},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13737. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As large language models (LLMs) are increasingly deployed, the generation of harmful content has become a critical safety concern. Existing safeguards operate at the input, output, or streaming-generation stages, while early-risk methods that rely on surface tokens or output logits may suffer from weak initial signals, and internals-based detectors using dense representations may retain highly entangled and redundant safety-irrelevant information. It therefore remains unclear whether the earliest post-generation hidden states already contain reliable signals about final-response harmfulness. To address this gap, we propose ForeSight, a first-token output-risk forecasting framework that distills weak and redundant early safety signals into compact, layer-aware risk representations. Experiments on five safety benchmarks and two target models demonstrate that ForeSight achieves superior and efficient early-risk forecasting while relying solely on first-token hidden states. The code is available at: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.13737 — 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 2026-10-15T22:51:09.834Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).