Metraipolis › archive › dc_21cfec20aae4

R2A-Sales: A Policy-Atom-Matched Benchmark for Rule-to-Action Gaps in Goal-Directed Sales Agents — arXiv:2610.34185

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.34185
License (as archived)
CC0-1.0
Archived
2026-10-27T07:57:28.789Z
SHA-256 of the archived text
6d1f428f1189964c6b1996a01cbd02f30ea3f6477d4a74958f6cc535dcd25833
Stable URL of this page
https://metraipolis.com/documento/dc_21cfec20aae4

How to cite

R2A-Sales: A Policy-Atom-Matched Benchmark for Rule-to-Action Gaps in Goal-Directed Sales Agents — arXiv:2610.34185 — arXiv (en), https://arxiv.org/abs/2610.34185 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_21cfec20aae4 (Metraipolis archive dc_21cfec20aae4, sha256 6d1f428f1189964c6b1996a01cbd02f30ea3f6477d4a74958f6cc535dcd25833)
BibTeX
@misc{dc21cfec20aae4,
  title = {R2A-Sales: A Policy-Atom-Matched Benchmark for Rule-to-Action Gaps in Goal-Directed Sales Agents — arXiv:2610.34185},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.34185},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_21cfec20aae4; sha256 6d1f428f1189964c6b1996a01cbd02f30ea3f6477d4a74958f6cc535dcd25833},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.34185. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM-based sales agents are often evaluated on static policy questions or task success, but deployment asks a sharper question: when a model can state a rule, will it choose compliant, tool-grounded actions under commercial pressure? We introduce R2A-Sales (Rule-to-Action), a policy-atom-matched benchmark for this question. Each item pairs a static probe with a multi-turn sales scenario sharing the same versioned policy atom, facts, evidence, behavioral constraints, and tool contract, so differences cannot be attributed to policy-content mismatch. In the interactive arm, a code-owned ledger verifies committed tool effects and terminal states; unsupported action claims receive no credit. We report static compliance $K$, grounded action compliance $A=\Pr(\text{safe terminal}\land\text{no hard violation}\land\text{non-echo})$, and the matched gap $G=K-A$. The frozen release spans three controlled Chinese-language sales domains, 24 policy atoms, 42 semantic cores, and 2,016 matched pairs. The confirmatory study audits 21,888 runs over 19 cores, covering eight backbones plus four Qwen3-8B prompt treatments (12 configurations). Static compliance is near ceiling $(K=89.9\%\text{--}99.1\%)$, yet eight-backbone action compliance is only $A=3.0\%\text{--}17.3\%$, giving $G=80.4\text{--}92.5$ percentage points. Since $A$ also requires safe termination, $G$ upper-bounds rule-to-action degradation rather than isolating pressure alone. No prompt-only treatment dominates all metrics. R2A-Sales shows that sales-agent reliability must be measured at committed action, not rule recital or task completion alone.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.34185 — 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 2026-10-27T07:55:30.681Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).