Metraipolis › archive › dc_da2860929fc0

AnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs — arXiv:2608.14320

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.14320
License (as archived): CC0-1.0
Archived: 2026-10-19T02:20:47.606Z
SHA-256 of the archived text: baa579439eda16b381918ead30493466808ee37c680eba15cea5705307fa084b
Stable URL of this page: https://metraipolis.com/documento/dc_da2860929fc0

How to cite

AnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs — arXiv:2608.14320 — arXiv (en), https://arxiv.org/abs/2608.14320 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_da2860929fc0 (Metraipolis archive dc_da2860929fc0, sha256 baa579439eda16b381918ead30493466808ee37c680eba15cea5705307fa084b)

BibTeX

@misc{dcda2860929fc0,
  title = {AnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs — arXiv:2608.14320},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14320},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_da2860929fc0; sha256 baa579439eda16b381918ead30493466808ee37c680eba15cea5705307fa084b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14320. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The anchoring effect is a cognitive bias in which an initial reference value shifts a later judgment toward itself. This effect is well established in human judgment and decision-making, and recent work suggests that large language models (LLMs) exhibit similar behavior. However, existing work on anchoring in LLMs typically evaluates only a narrow set of anchor pathways and rarely distinguishes irrelevant from plausible anchors. We introduce AnchorBench, a benchmark for the anchoring effect in LLMs that evaluates multiple anchor pathways under an explicit anchor relevance axis. Across fourteen models, including ten open-weight models and four frontier API models, and a large set of controlled prompts, we find that (1) anchoring is strongly pathway-dependent, (2) plausible anchors usually induce larger shifts than irrelevant ones when introduced through stronger pathways, (3) anchor influence generally weakens as the anchor moves farther from the evidence-supported answer, most clearly on External and RAG, and (4) high task accuracy on the anchor-free control condition (Acc$_{10}$: answers within 10 points of gold) does not guarantee robustness: even frontier API models above 95% control accuracy remain susceptible to plausible anchors.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.14320 — 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 2026-10-19T02:02:02.805Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).