arXiv cs.CL· Mochamad Rizal Fauzan, Naila Nursyifa, Irgi Surya·· 11 小时前AI 评分18
Halal-NeSyNet：子句级神经符号交叉注意力用于食品配料审计与清真物流风险分级
Halal-NeSyNet: Clause-Scoped Neuro-Symbolic Cross-Attention for Automated Food Ingredient Auditing and Risk-Aware Halal Logistics Triage
AI 导读
Halal-NeSyNet 将 DistilBERT 编码器与 16 维规则符号向量通过交叉注意力和门控融合结合，在 3,199 条配料审计基准上达到 97.04±0.34% 准确率和 99.67±0.50% 反事实三元组一致性，而纯文本 TF-IDF 基线仅 86.25-86.71%，假清真泄漏率 13.42-21.73%。
正文 · 原文
Abstract:Verifying that a packaged food is Halal requires reading ingredient declarations and E-number additives before goods enter certified warehouses. Dual-origin additives such as E471, E422, E441 and E920 are Mushbooh (doubtful) unless a botanical, microbial or certified-Halal qualifier appears in the same ingredient clause, so a plant-based qualifier on one clause must not clear an unqualified additive in another. We call a failure to respect this scope cross-clause modifier confounding and build a benchmark of 3,199 items (277 Open Food Facts records and 2,922 template-generated clause-swap items, labelled by an explicit 50-code rule base rather than by a certification authority) with a 540-item counterfactual suite. We propose Halal-NeSyNet, which couples a DistilBERT encoder with a 16-dimensional rule-derived symbolic vector through cross-attention and gated fusion, trained with an asymmetric false-Halal risk loss. Over five seeds, text-only TF-IDF baselines reach 86.25-86.71 % accuracy with 13.42-21.73 % false-Halal leakage and 6.33-26.00 % counterfactual triad consistency, whereas Halal-NeSyNet reaches 97.04+/-0.34 % and 99.67+/-0.50 %. Gradient boosting given the same symbolic vector reaches 96.25+/-1.24 % (+0.59 points from Halal-NeSyNet, 95 % CI [-1.11, +2.28]), so most of the gain comes from clause-scoped symbolic features and not from the network, and the risk loss did not help. Holding out whole template families, text-only models fall to 61.91-78.35 %, Halal-NeSyNet to 90.87+/-11.76 % and logistic regression with the symbolic vector to 91.11+/-2.88 %, with far more spread for Halal-NeSyNet. A three-lane receiving simulation auto-clears 49.33 % of cargo at 0.76 % green-lane contamination. Because labels and features come from one rule engine and 91.3 % of the items are templated, the results measure fidelity to a rule policy, not religious certification.
| Comments: | 14 pages, 7 figures, 13 tables |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.01612 [cs.CL] |
| (or arXiv:2611.01612v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.01612 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mochamad Rizal Fauzan [view email]
[v1]
Sun, 1 Nov 2026 05:56:41 UTC (151 KB)
来源：arXiv cs.CL · arxiv.org