@inproceedings{li-etal-2026-cost,
title = "Cost-{LLM}: Efficient Multimodal Procurement-Cost Reasoning for {E}-Commerce Price Governance",
author = "Li, Zhiyin and
Huang, Fei and
Kong, Tengteng and
Jie, Tang and
Li, Shengjie",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.13/",
pages = "199--210",
ISBN = "979-8-89176-460-6",
abstract = "Estimation of product procurement cost is a key capability for price governance on large-scale e-commerce platforms: from images, structured attributes, comparable products, and pricing knowledge, a practical system must infer a reasonable cost to surface suspiciously inflated or potentially non-compliant listings. Directly invoking general-purpose VLMs is prohibitive at high QPS (queries per second), and their verbose rationales complicate auditing and rule integration. We present Cost-LLM, a compact 7B multimodal cost-estimation model trained by a two-stage post-training pipeline: segment-normalized SFT on progressive pricing rationales, followed by GRPO with an asymmetric MAPE-based reward aligned with cost accuracy. On an in-house benchmark covering 105 fine-grained categories, Cost-LLM reaches MAPE 29.47{\%} and Acc@20 40.42{\%}, yielding the best point estimates among the evaluated systems. Deployed as an auxiliary signal in a price-anomaly pipeline, it provides a 30.73{\%} relative complementary-detection yield over the production baseline at an audited precision-matched operating point. Its locally deployed serving latency is 0.4-0.8 s per listing, compared with 3-8 s observed for external commercial LLM/VLM APIs. We share the design choices, training recipe, and deployment lessons."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-cost">
<titleInfo>
<title>Cost-LLM: Efficient Multimodal Procurement-Cost Reasoning for E-Commerce Price Governance</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhiyin</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fei</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tengteng</namePart>
<namePart type="family">Kong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tang</namePart>
<namePart type="family">Jie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shengjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>Estimation of product procurement cost is a key capability for price governance on large-scale e-commerce platforms: from images, structured attributes, comparable products, and pricing knowledge, a practical system must infer a reasonable cost to surface suspiciously inflated or potentially non-compliant listings. Directly invoking general-purpose VLMs is prohibitive at high QPS (queries per second), and their verbose rationales complicate auditing and rule integration. We present Cost-LLM, a compact 7B multimodal cost-estimation model trained by a two-stage post-training pipeline: segment-normalized SFT on progressive pricing rationales, followed by GRPO with an asymmetric MAPE-based reward aligned with cost accuracy. On an in-house benchmark covering 105 fine-grained categories, Cost-LLM reaches MAPE 29.47% and Acc@20 40.42%, yielding the best point estimates among the evaluated systems. Deployed as an auxiliary signal in a price-anomaly pipeline, it provides a 30.73% relative complementary-detection yield over the production baseline at an audited precision-matched operating point. Its locally deployed serving latency is 0.4-0.8 s per listing, compared with 3-8 s observed for external commercial LLM/VLM APIs. We share the design choices, training recipe, and deployment lessons.</abstract>
<identifier type="citekey">li-etal-2026-cost</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.13/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>199</start>
<end>210</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Cost-LLM: Efficient Multimodal Procurement-Cost Reasoning for E-Commerce Price Governance
%A Li, Zhiyin
%A Huang, Fei
%A Kong, Tengteng
%A Jie, Tang
%A Li, Shengjie
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F li-etal-2026-cost
%X Estimation of product procurement cost is a key capability for price governance on large-scale e-commerce platforms: from images, structured attributes, comparable products, and pricing knowledge, a practical system must infer a reasonable cost to surface suspiciously inflated or potentially non-compliant listings. Directly invoking general-purpose VLMs is prohibitive at high QPS (queries per second), and their verbose rationales complicate auditing and rule integration. We present Cost-LLM, a compact 7B multimodal cost-estimation model trained by a two-stage post-training pipeline: segment-normalized SFT on progressive pricing rationales, followed by GRPO with an asymmetric MAPE-based reward aligned with cost accuracy. On an in-house benchmark covering 105 fine-grained categories, Cost-LLM reaches MAPE 29.47% and Acc@20 40.42%, yielding the best point estimates among the evaluated systems. Deployed as an auxiliary signal in a price-anomaly pipeline, it provides a 30.73% relative complementary-detection yield over the production baseline at an audited precision-matched operating point. Its locally deployed serving latency is 0.4-0.8 s per listing, compared with 3-8 s observed for external commercial LLM/VLM APIs. We share the design choices, training recipe, and deployment lessons.
%U https://aclanthology.org/2026.emnlp-industry.13/
%P 199-210
Markdown (Informal)
[Cost-LLM: Efficient Multimodal Procurement-Cost Reasoning for E-Commerce Price Governance](https://aclanthology.org/2026.emnlp-industry.13/) (Li et al., EMNLP 2026)
ACL