Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track

SAGE: A Generic Framework for LLM Safety Evaluation (2025.emnlp-industry)

Challenge:	Recent advances in large language models have improved the detection of non-compliant content, but critical gaps persist in fine-grained understanding, explainability, and generalization.
Approach:	They propose a framework that combines active reinforcement learning, fine-grained violation understanding and progressive multi-stage training.
Outcome:	The proposed framework outperforms general-purpose LLMs and specialized models in fine-grained violation understanding, explainability, and generalization.

Copied to clipboard

Madhur Jindal, Hari Shrawgi, Parag Agrawal, Sandipan Dandapat

Challenge:	Current safety evaluation methodologies focus on single-turn interactions with generic policies, failing to capture conversational dynamics of real-world usage and application-specific harms.
Approach:	They propose a framework for customized and dynamic harm evaluations that employs prompted adversarial agents with diverse personalities based on the Big Five model.
Outcome:	The proposed framework enables system-aware multi-turn conversations that adapt to target applications and harm policies.

CRAB: A Benchmark for Evaluating Curation of Retrieval-Augmented LLMs in Biomedicine (2025.emnlp-industry)

Copied to clipboard

Hanmeng Zhong, Linqing Chen, Wentao Wu, Weilei Wang

VENUS: A VLLM-driven Video Content Discovery System for Real Application Scenarios (2025.emnlp-industry)

Challenge:	Recent development in Retrieval-Augmented Large Language Models (LLMs) have shown great promise in biomedical applications.
Approach:	They propose a multilingual benchmark to evaluate retrieval-augmented large language models' curation ability.
Outcome:	The proposed benchmark is available in English, French, German and Chinese.

Copied to clipboard

Minyi Zhao, Yi Liu, Jianfeng Wen, Boshen Zhang, Hailang Chang, Zhiheng Ouyang, Jie Wang, Wensong He, Shuigeng Zhou

Visual Data in NLP

Challenge:	Video Content Discovery (VCD) is to identify specific videos defined by a pre-specified text policy.
Approach:	They propose a Vision-Language Large Model-driven video content discovery system called VENUS to solve these problems.
Outcome:	The proposed system generates high-quality, VCD-specific data for model training and extends it to support it better.

FT-MDT: Extracting Decision Trees from Medical Texts via a Novel Low-rank Adaptation Method (2025.emnlp-industry)

Copied to clipboard

Yuheng Li, Jiechao Gao, Wei Han, Wenwen Ouyang, Wei Zhu, Hui Yi Leong

PolyNorm: Few-Shot LLM-Based Text Normalization for Text-to-Speech (2025.emnlp-industry)

Challenge:	Existing methods for extracting medical decision trees rely on manual annotation . PI-LoRA is a low-rank adaptation method for extract medical decision tree from clinical guidelines and textbooks .
Approach:	They propose a low-rank adaptation method for automatically extracting medical decision trees from clinical guidelines and textbooks.
Outcome:	The proposed method outperforms existing methods for the Text2MDT task while maintaining a lightweight architecture.

Copied to clipboard

Michel Wong, Ali Alshehri, Sophia Kao, Haotian He

Speech & Audio in NLP

Challenge:	Text Normalization (TN) is a key preprocessing step in Text-to-Speech systems.
Approach:	They propose a prompt-based approach to TN using Large Language Models (LLMs) they propose scalable experimentation across languages to reduce the reliance on manual rules .
Outcome:	The proposed approach reduces the reliance on manual rules and enables broader linguistic applicability with minimal human intervention across eight languages.

Audio Query Handling System with Integrated Expert Models and Contextual Understanding (2025.emnlp-industry)

Copied to clipboard

Naveen Vakada, Arvind Krishna Sridhar, Yinyi Guo, Erik Visser

Generative Reviewer Agents: Scalable Simulacra of Peer Review (2025.emnlp-industry)

Challenge:	Existing chatbots are limited to specific audio tasks, but the domain of audio content related queries remains underexplored.
Approach:	They propose to use an intent classifier to route queries to audio-related experts using a diverse audio query dataset.
Outcome:	The proposed system outperforms state-of-the-art LLMs on custom audio tasks and MMAU sound set benchmarks.

Copied to clipboard

Nicolas Bougie, Narimawa Watanabe

Challenge:	Existing peer review mechanisms are limited by the small fraction of researchers with established networks.
Approach:	They propose a system that extends a large language model and equips agents with reviewer personas derived from historical data to enable generative reviewers.
Outcome:	The proposed architecture performs comparable to human reviewers in providing detailed feedback and predicting paper outcomes.

Aligning LLMs for Multilingual Consistency in Enterprise Applications (2025.emnlp-industry)

Copied to clipboard

Amit Agarwal, Hansa Meghwani, Hitesh Laxmichand Patel, Tao Sheng, Sujith Ravi, Dan Roth

RCI: A Score for Evaluating Global and Local Reasoning in Multimodal Benchmarks (2025.emnlp-industry)

Challenge:	Large language models (LLMs) remain unreliable for global enterprise applications due to performance gaps between high-resource and mid/low-resourced languages .
Approach:	They propose a batch-wise alignment strategy that aligns model outputs across languages . this method improves non-English accuracy by up to 23.9% without compromising English performance .
Outcome:	The proposed approach improves non-English accuracy by up to 23.9% without compromising English performance, model reasoning, or retrieval quality.

Copied to clipboard

Amit Agarwal, Hitesh Laxmichand Patel, Srikant Panda, Hansa Meghwani, Jyotika Singh, Karan Dua, Paul Li, Tao Sheng, Sujith Ravi, Dan Roth

LP Data Pipeline: Lightweight, Purpose-driven Data Pipeline for Large Language Models (2025.emnlp-industry)

Challenge:	Existing evaluation methods do not explicitly measure this distinction, hindering effective dataset curation and real-world focused model development.
Approach:	They introduce a region-based score to quantify a dataset's reliance on global versus local visual information.
Outcome:	The proposed model-based score systematically compares model performance on image patches versus full images to determine if tasks require holistic image understanding or can be solved with partial or localized visual cues.

Copied to clipboard

Yungi Kim, Hyunsoo Ha, Seonghoon Yang, Sukyung Lee, Jihoo Kim, Chanjun Park

Toward Reliable Clinical Coding with Language Models: Verification and Lightweight Adaptation (2025.emnlp-industry)

Challenge:	Creating high-quality datasets for large language models often relies on resource-intensive, GPU-accelerated models for quality filtering, making the process time-consuming and costly.
Approach:	They propose a framework that operates entirely on CPUs to streamline the processes of dataset extraction, filtering, and curation.
Outcome:	The proposed framework reduces preparation time and costs while maintaining high data quality while enhancing the applicability of LLMs in specialized contexts.

Copied to clipboard

Moy Yuan, Han-Chin Shing, Mitch Strong, Chaitanya Shivade

Enhancing Talent Search Ranking with Role-Aware Expert Mixtures and LLM-based Fine-Grained Job Descriptions (2025.emnlp-industry)

Challenge:	Existing methods for clinical code verification fail to account for hierarchical misalignments . standardized coding systems such as ICD-10-CM1 ensure consistency across medical records.
Approach:	They propose to use prompt engineering and small-scale fine-tuning to improve accuracy without the computational overhead of search-based methods.
Outcome:	The proposed task is a standalone task and a pipeline component to address hierarchical near-miss errors without the computational overhead of search-based methods.

Copied to clipboard

Jihang Li, Bing Xu, Zulong Chen, Chuanfei Xu, Minping Chen, Suyu Liu, Ying Zhou, Zeyi Wen

PCRI: Measuring Context Robustness in Multimodal Models for Enterprise Applications (2025.emnlp-industry)

Challenge:	Existing talent search approaches fail to capture nuanced job-specific preferences and mitigate noise from subjective human judgments.
Approach:	They propose a framework that extracts fine-grained recruitment signals from job descriptions and historical hiring data and employs a role-aware multi-gate MoE network to capture behavioral differences across recruiter roles.
Outcome:	The proposed framework improves talent search effectiveness and delivers substantial business value.

Copied to clipboard

Hitesh Laxmichand Patel, Amit Agarwal, Srikant Panda, Hansa Meghwani, Karan Dua, Paul Li, Tao Sheng, Sujith Ravi, Dan Roth

CitySim: Modeling Urban Behaviors and City Dynamics with Large-Scale LLM-Driven Agent Simulation (2025.emnlp-industry)

Challenge:	Existing evaluation metrics for Multimodal Large Language Models (MLLMs) are inadequate to assess their robustness to irrelevant or distracting visual context.
Approach:	They propose a patch-context-robustness index to measure MLLMs' robustness to visual context variations.
Outcome:	The proposed score measures the robustness of MLLMs to visual contexts across 15 vision-language benchmarks.

Copied to clipboard

Nicolas Bougie, Narimawa Watanabe

Challenge:	Existing models rely on rigid, hand-crafted rules to model nuanced behavior in urban environments.
Approach:	They propose an urban simulator that generates realistic daily schedules using a recursive value-driven approach that balances mandatory activities, personal habits, and situational factors.
Outcome:	The proposed urban simulator exhibits closer alignment with real humans than previous work.

Evaluating Conversational Agents with Persona-driven User Simulations based on Large Language Models: A Sales Bot Case Study (2025.emnlp-industry)

Copied to clipboard

Justyna Gromada, Alicja Kasicka, Ewa Komkowska, Lukasz Krajewski, Natalia Krawczyk, Morgan Veyret, Bartosz Przybył, Lina M. Rojas-Barahona, Michał K. Szczerbak

Mirror in the Model: Ad Banner Image Generation via Reflective Multi-LLM and Multi-modal Agents (2025.emnlp-industry)

Challenge:	Recent advances in LLMs enable sophisticated user simulations that can replace traditional rule-based evaluations.
Approach:	They propose a persona-driven approach to conversational agent evaluation using Large Language Models (LLMs) they introduce a dataset of customer personas, which are then used to configure a single LLM-based user simulator.
Outcome:	The proposed model emulates nuanced customer roles and can implement cross-selling strategies with minimal impact on customer satisfaction, varying by customer type.

Copied to clipboard

Zhao Wang, Bowen Chen, Yotaro Shimose, Sota Moriyama, Heng Wang, Shingo Takamatsu

Leveraging Product Catalog Patterns for Multilingual E-commerce Product Attribute Prediction (2025.emnlp-industry)

Challenge:	Recent advances in generative modeling have greatly improved image synthesis quality.
Approach:	They propose an agentic refinement framework for automatic ad banner generation that integrates a hierarchical multimodal agent system with a coordination loop.
Outcome:	The proposed model outperforms existing models in real-world banner design scenarios.

Copied to clipboard

Bryan Zhang, Suleiman A. Khan, SteCphan Walter

ECom-Bench: Can LLM Agent Resolve Real-World E-commerce Customer Support Issues? (2025.emnlp-industry)

Challenge:	E-commerce stores increasingly use Large Language Models to improve catalog data quality . a critical challenge is accurately predicting missing structured attribute values .
Approach:	They propose a retrieval-augmented system that leverages existing product catalog entries to guide LLM predictions for missing attributes.
Outcome:	The proposed system improves catalog data quality by 34% and accuracy by 0.8% . the proposed model can predict missing attributes in multilingual product catalogs .

Copied to clipboard

Haoxin Wang, Xianhan Peng, Huang Cheng, Yizhe Huang, Ming Gong, Chenghan Yang, Yang Liu, Jiang Lin

Challenge:	ECom-Bench is a benchmark framework for evaluating LLM agent with multimodal capabilities in e-commerce customer support domain.
Approach:	They introduce a benchmark framework for evaluating LLM agent with multimodal capabilities in the e-commerce customer support domain.
Outcome:	The proposed benchmark features dynamic user simulation based on persona information from real e-commerce customer interactions and a realistic task dataset derived from authentic ecommerce dialogues.

ProCut: LLM Prompt Compression via Attribution Estimation (2025.emnlp-industry)

Copied to clipboard

Zhentao Xu, Fengyi Li, Albert C. Chen, Xiaofeng Wang

A Reasoner for Real-World Event Detection: Scaling Reinforcement Learning via Adaptive Perplexity-Aware Sampling Strategy (2025.emnlp-industry)

Challenge:	ProCut compresses prompts using attribution analysis to reduce prompt size and latency.
Approach:	They propose a framework that compresses prompts through attribution analysis using a heuristic and attribution-based attribution model.
Outcome:	The proposed framework reduces prompt size by 78% while maintaining or improving task performance by 62%.

Copied to clipboard

Xiaoyun Zhang, Jingqing Ruan, Xing Ma, Yawen Zhu, Jiansong Chen, Ke Zeng, Xunliang Cai

Detecting Omissions in LLM-Generated Medical Summaries (2025.emnlp-industry)

Challenge:	Existing methods for abnormal event detection face two predominant limitations . existing methods rely on specialized small models and are limited by performance bottlenecks .
Approach:	They propose a framework that leverages the advanced reasoning capabilities of large language models for abnormal event detection.
Outcome:	The proposed framework achieves the highest F1 score and an average improvement of 9.59% in OOD transfer tests.

Copied to clipboard

Achir Oukelmoun, Nasredine Semmar, Gaël de Chalendar, Clement Cormi, Mariame Oukelmoun, Eric Vibert, Marc-Antoine Allard

LEAF: Learning and Evaluation Augmented by Fact-Checking to Improve Factualness in Large Language Models (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) have created a number of use cases in the medical field . omissions in summaries can jeopardize the decision-making process .
Approach:	They propose a dataset to evaluate omissions in large-scale medical summaries . they propose 'embedKDECheck' method that uses embeddings generated by a third-party NLP model .
Outcome:	The proposed method is well-suited for resource-constrained environments.

Copied to clipboard

Hieu Tran, Junda Wang, Yujan Ting, Hong Yu, Weijing Huang, Terrence Chen

Ethical NLP

Challenge:	Large language models (LLMs) struggle with factual accuracy in knowledge-intensive domains like healthcare.
Approach:	They propose a framework for improving LLM factuality in medical question answering . RAFE, Fact-Check-then-RAG and Learning from Fact Check are components .
Outcome:	Experimental results show that LEAF outperforms Factcheck-GPT in detecting inaccuracies and corrects errors without labeling . the framework provides a scalable solution for industrial applications requiring high factuality scores.

ReAct Meets Industrial IoT: Language Agents for Data Access (2025.emnlp-industry)

Copied to clipboard

James T Rayfield, Shuxin Lin, Nianjun Zhou, Dhaval C Patel

ProductAgent: Benchmarking Conversational Product Search Agent with Asking Clarification Questions (2025.emnlp-industry)

Challenge:	a framework for domain-specific language agents is being developed for industrial automation . a novel approach to adapting these systems to domain-based applications poses new challenges .
Approach:	They propose a framework for deploying domain-specific language agents that can query industrial sensor data using natural language.
Outcome:	The proposed framework outperforms standard prompting baselines across multiple LLMs including smaller models.

Copied to clipboard

Jingheng Ye, Yong Jiang, Xiaobin Wang, Yinghui Li, Yangning Li, Pengjun Xie, Fei Huang

MADS: Multi-Agent Dialogue Simulation for Diverse Persuasion Data Generation (2025.emnlp-industry)

Challenge:	Recent advances in conversational information seeking (CIS) suggest a remedy for the lack of interactive clarification when people face unfamiliar domains.
Approach:	They propose a fully autonomous conversational information-seeking agent that couples large language models with a set of domain-specific tools to provide product demand clarification.
Outcome:	The proposed agent can iterate over 2,000 automatically generated sessions and score high on real-world evaluations without human annotation.

Copied to clipboard

Mingjin Li, Yu Liu, Huayi Liu, Xiang Ye, Chao Jiang, Hongguang Zhang, Yu Ruan

On-device System of Compositional Multi-tasking in Large Language Models (2025.emnlp-industry)

Challenge:	Recent studies show that LLM-based agents exhibit superior moral and emotional language performance compared to humans, raising expectations for their deployment in persuasive tasks.
Approach:	They propose a framework for generating persuasive multi-turn dialogues via agent self-play using user agents designed to simulate diverse persona-driven behaviors, a Dialog Agent executing task-oriented persuasion strategies and an Optimization Agent evaluating and refining dialogue outcomes.
Outcome:	The proposed framework significantly improved the persuasion capacity of small LLMs, increasing the organic traffic conversion rate by 22.4% (from 1.83% to 2.24%) .

Copied to clipboard

Ondrej Bohdal, Konstantinos Theodosiadis, Asterios Mpatziakas, Dimitrios Filippidis, Iro Spyrou, Christos Zonios, Anastasios Drosou, Dimosthenis Ioannidis, Kyenghun Lee, Jijoong Moon, Hyeonmok Ko, Mete Ozay, Umberto Michieli

Select-then-Route : Taxonomy guided Routing for LLMs (2025.emnlp-industry)

Challenge:	Existing approaches to generative AI for large language models struggle when executing complex tasks simultaneously.
Approach:	They propose a novel approach tailored specifically for compositional multi-tasking scenarios . they add a learnable projection layer on top of the combined summarization and translation adapters.
Outcome:	The proposed approach performs well and is fast in both cloud-based and on-device implementations.

Copied to clipboard

Soham Shah, Kumar Shridhar

Challenge:	Large language models have boosted performance across a broad spectrum of tasks . sending each query to the most suitable model is prohibitively expensive .
Approach:	They propose a framework that selects a small pool of LLMs and routes queries through an adaptive cascade.
Outcome:	The proposed framework improves accuracy and latency by 4X while reducing inference cost.

FABRIC: Fully-Automated Broad Intent Categorization in E-commerce (2025.emnlp-industry)

Copied to clipboard

Anna Tigunova, Philipp Schmidt, Damla Ezgi Akcora

MKT: A Multi-Stage Knowledge Transfer Framework to Mitigate Catastrophic Forgetting in Multi-Domain Chinese Spelling Correction (2025.emnlp-industry)

Challenge:	Existing query classification models have excellent predictive performance on single-intent queries, but there is little research on predicting multiple-intentions for broad queries.
Approach:	They propose to combine user click data, query-item relevance and LLM judgments to create an automatic method for multi-label e-commerce query classification.
Outcome:	The proposed method reduces the ambiguity of the annotations by blending the label assessment from three different sources: user click data, query-item relevance and LLM judgments.

Copied to clipboard

Peng Xing, Yinghui Li, Shirong Ma, Xinnian Liang, Haojing Huang, Yangning Li, Shu-Yu Guo, Hai-Tao Zheng, Wenhao Jiang, Ying Shen

End-to-End Aspect-Guided Review Summarization at Scale (2025.emnlp-industry)

Challenge:	Chinese Spelling Correction (CSC) is a model that detects and corrects spelling errors in given sentences.
Approach:	They propose a model-agnostic model with an evolving teacher model and dynamic distillation weights for knowledge transfer in each domain rather than focusing solely on new domain knowledge.
Outcome:	The proposed model-agnostic framework is based on an evolving teacher model and dynamic distillation weights for knowledge transfer in each domain, rather than focusing solely on new domain knowledge.

Copied to clipboard

Ilya Boytsov, Vinny DeGenova, Mikhail Balyasin, Joseph Walt, Caitlin Eusden, Marie-Claire Rochat, Margaret Pierson

Summarization

Challenge:	Existing methods to generate concise product review summaries are prone to hallucination, omission of important facts, and factual errors.
Approach:	They propose a large language model-based system that combines aspect-based sentiment analysis with guided summarization to generate concise product review summaries.
Outcome:	The proposed system generates concise and interpretable product review summaries using a large language model (LLM) dataset.

SLOT: Structuring the Output of Large Language Models (2025.emnlp-industry)

Copied to clipboard

Zhengyuan Shen, Darren Yow-Bang Wang, Soumya Smruti Mishra, Zhichao Xu, Yifei Teng, Haibo Ding

QuackIR: Retrieval in DuckDB and Other Relational Database Management Systems (2025.emnlp-industry)

Challenge:	Structured outputs are essential for large language models (LLMs) but often deviate from predefined schemas hampering reliable application development.
Approach:	They propose a model-agnostic approach that transforms unstructured LLM outputs into precise structured formats.
Outcome:	The proposed model-agnostic approach transforms unstructured LLM outputs into precise structured formats.

Copied to clipboard

Yijun Ge, Zijian Chen, Jimmy Lin

Benchmarking Deep Search over Heterogeneous Enterprise Data (2025.emnlp-industry)

Challenge:	Existing vector databases for RAG are needed for large language models, but there are no alternatives.
Approach:	They propose to leverage existing relational databases for retrieval-augmented generation . they use duckDB, SQLite, and PostgreSQL integrations to demonstrate their effectiveness .
Outcome:	The proposed approach is comparable to existing IR tools.

Copied to clipboard

Prafulla Kumar Choubey, Xiangyu Peng, Shilpa Bhagavath, Kung-Hsiang Huang, Caiming Xiong, Chien-Sheng Wu

RLHF Algorithms Ranked: An Extensive Evaluation Across Diverse Tasks, Rewards, and Hyperparameters (2025.emnlp-industry)

Challenge:	Existing methods struggle to conduct deep searches and retrieve all necessary evidence.
Approach:	They propose a benchmark for evaluating deep search, a retrieval-augmented generation that requires source-aware, multi-hop reasoning over diverse, sparsed, but related sources.
Outcome:	The proposed benchmarks show that even the best-performing agentic RAG methods achieve an average performance score of 32.96 on the benchmark.

Copied to clipboard

Lucas Spangher, Rama Kumar Pasumarthi, Nick Masiewicki, William F. Arnold, Aditi Kaushal, Dale Johnson, Peter Grabowski, Eugene Ie

Challenge:	Proximal Policy Optimization (PPO) has fallen out of favor for Large Language Models (LLMs), but its complexity and inefficiency have spurred the investigation of simpler alternatives.
Approach:	They evaluate 17 RLHF algorithms on two benchmarks, OpenAI’s TL;DR Summarization and Anthropic’s Helpfulness / Harmlessness.
Outcome:	The proposed methods are based on OpenAI’s TL;DR Summarization and Anthropic’s Helpfulness / Harmlessness benchmarks with two different reward models and a Rules based reward model.

Predicting Cross-lingual Trends in Microblogs (2025.emnlp-industry)

Copied to clipboard

Satoshi Akasaki

Generating Fine Details of Entity Interactions (2025.emnlp-industry)

Challenge:	Existing trend prediction methods only make predictions within a language, but this is not enough to predict cross-lingual trends.
Approach:	They propose a method to predict which microblog trends will cross linguistic boundaries to become popular in other languages and when.
Outcome:	The proposed model outperforms existing trend prediction methods and LLM-based approaches by 4% in F1-score .

Copied to clipboard

Xinyi Gu, Jiayuan Mao

AutoCVSS: Assessing the Performance of LLMs for Automated Software Vulnerability Scoring (2025.emnlp-industry)

Challenge:	Existing text-to-image models excel at generating high-quality object-centric images from instructions, but lack of data for complex interactions.
Approach:	They propose a multimodal Large Language Models-generated dataset to benchmark and enhance interaction-rich images.
Outcome:	The proposed approach improves image quality and automatic and human evaluations show improvements.

Copied to clipboard

Davide Sanvito, Giovanni Arriciati, Giuseppe Siracusano, Roberto Bifulco, Michele Carminati

SFAL: Semantic-Functional Alignment Scores for Distributional Evaluation of Auto-Interpretability in Sparse Autoencoders (2025.emnlp-industry)

Challenge:	Increasing number of daily disclosed software vulnerabilities imposes significant pressure on security analysts, extending the time between disclosure and exploitation.
Approach:	They propose to use Large Language Models to automate vulnerability risk score prediction using the industrial CVSS standard.
Outcome:	The proposed model complements baselines in data-scarce settings without annotated data, highlighting their value in improving vulnerability management.

Copied to clipboard

Fabio Mercorio, Filippo Pallucchini, Daniele Potertì, Antonio Serino, Andrea Seveso

Just One is Enough: An Existence-based Alignment Check for Robust Japanese Pronunciation Estimation (2025.emnlp-industry)

Challenge:	Interpreting the internal representations of large language models (LLMs) is crucial for their deployment in real-world applications, impacting areas such as AI safety, debugging, and compliance.
Approach:	They propose an alternative evaluation strategy that assesses the alignment between the semantic neighbourhoods of features and their functional neighbourhoods by using co-occurrence statistics.
Outcome:	The proposed evaluation strategy reduces reliance on scoring on large-scale models and improves efficiency and cost-effectiveness.

Copied to clipboard

Hayate Nakano, Nobuhiro Kaji

Towards Enforcing Company Policy Adherence in Agentic Workflows (2025.emnlp-industry)

Challenge:	Existence-based alignment has been used to detect pronunciation errors in Japanese NLP, but finding reliable attention heads remains challenging.
Approach:	They propose a method that detects and corrects pronunciation errors in Japanese by using beam search.
Outcome:	The proposed method reduces hallucinations and omissions and improves pronunciation estimation by over 2.5%.

Copied to clipboard

Naama Zwerdling, David Boaz, Ella Rabinovich, Guy Uziel, David Amid, Ateret Anaby Tavor

Challenge:	Large Language Models (LLMs) agents are transforming business processes with minimal human oversight.
Approach:	They propose a deterministic, transparent, and modular framework for enforcing business policy adherence in agentic workflows.
Outcome:	The proposed framework shows encouraging preliminary results in policy enforcement on the -bench Airlines domain.

Learning to Translate Ambiguous Terminology by Preference Optimization on Post-Edits (2025.emnlp-industry)

Copied to clipboard

Nathaniel Berger, Johannes Eschbach-Dymanus, Miriam Exel, Matthias Huck, Stefan Riezler

Machine Translation

Challenge:	Ambiguous terminology can make translation difficult, especially in corporate contexts.
Approach:	They propose to learn how to disambiguate terminology based on human post-edits . they use preference optimization to optimize for correctness using the term post-Edit .
Outcome:	The proposed framework improves term accuracy over a translation oriented LLM without significant losses in COMET score.

More Data or Better Data? A Critical Analysis of Data Selection and Synthesis for Mathematical Reasoning (2025.emnlp-industry)

Copied to clipboard

Yike Zhao, Simin Guo, Ziqing Yang, Shifan Han, Dahua Lin, Fei Tan

SRS-Stories: Vocabulary-constrained multilingual story generation for language learning (2025.emnlp-industry)

Challenge:	Despite various proposed data construction methods, their practical utility in real-world pipelines remains underexplored.
Approach:	They conduct a comprehensive analysis of open-source datasets and data synthesis techniques for mathematical reasoning under a unified pipeline designed to mirror training and deployment scenarios.
Outcome:	The proposed pipelines mirror training and deployment scenarios and are suitable for industrial applications.

Copied to clipboard

Wiktor Kamzela, Mateusz Lango, Ondrej Dusek

Banking Done Right: Redefining Retail Banking with Language-Centric AI (2025.emnlp-industry)

Challenge:	Existing methods for learning foreign languages are to use a spaced repetition system to learn new vocabulary.
Approach:	They use large language models to generate personalized stories using only the vocabulary they know.
Outcome:	The generated stories are more grammatical, coherent, and provide better examples of word usage than the standard beam search approach.

Copied to clipboard

Xin Jie Chua, Jeraelyn Ming Li Tan, Jia Xuan Tan, Soon Chang Poh, Yi Xian Goh, Debbie Hui Tian Choong, Foong Chee Mun, Sze Jue Yang, Chee Seng Chan

Graph of Attacks with Pruning: Optimizing Stealthy Jailbreak Prompt Generation for Enhanced LLM Content Moderation (2025.emnlp-industry)

Challenge:	This is the first global regulator-approved deployment where conversational AI functions as the primary banking interface.
Approach:	They propose a framework that powers a conversational AI framework that is powered by a closed-source LLM developed internally and replaces rigid multi-screen workflows with a single dialogue orchestrated by four LLM-powered agents.
Outcome:	The proposed framework replaces multi-screen workflows with a single dialogue orchestrated by four LLM-powered agents (Guardrails, Intent, Payment, and FAQ).

Copied to clipboard

Daniel Schwartz, Dmitriy Bespalov, Zhe Wang, Ninad Kulkarni, Yanjun Qi

Structural Reward Model: Enhancing Interpretability, Efficiency, and Scalability in Reward Modeling (2025.emnlp-industry)

Challenge:	Existing jailbreaking methods create adversarial prompts to bypass LLM safeguards.
Approach:	They propose a framework for generating stealthy jailbreak prompts that enables knowledge sharing across attack paths.
Outcome:	The proposed framework outperforms state-of-the-art methods for attacking both open and closed LLMs with attack success rates of >96%.

Copied to clipboard

Xiaoyu Liu, Di Liang, Hongyu Shan, Peiyang Liu, Yonghao Liu, Muling Wu, Yuntao Li, Xianjie Wu, Li Miao, Jiangrong Shen, Minlong Peng

Controllable Clustering with LLM-driven Embeddings (2025.emnlp-industry)

Challenge:	Generative RMs (GRMs) lack contextual and background information during inference, leading to incomplete evaluations.
Approach:	They propose a modular and interpretable framework that integrates side-branch models as auxiliary feature generators.
Outcome:	The proposed framework outperforms scalar and saline reward models in robustness and alignment with human preferences.

Copied to clipboard

Kerria Pang-Naylor, Shivani Manivasagan, Aitong Zhong, Mehak Garg, Nicholas Mondello, Blake Buckner, Jonathan P. Chang, Khyati Mahajan, Masoud Hashemi, Fabio Casati

Text Clustering

Challenge:	Unsupervised text clustering is unlikely to produce groupings that work across use cases . authors present techniques to effectively control text embeddings with minimal human input .
Approach:	They propose techniques to control text embeddings with minimal human input . they evaluate clustering performance for datasets with multiple independent labels .
Outcome:	The proposed techniques improve clustering for one perspective or use case, but at a tradeoff in performance for another use case.

SpeechLLMs for Large-scale Contextualized Zero-shot Slot Filling (2025.emnlp-industry)

Copied to clipboard

Kadri Hacioglu, Manjunath K E, Andreas Stolcke

NurseLLM: The First Specialized Language Model for Nursing (2025.emnlp-industry)

Challenge:	Slot filling is a key subtask in spoken language understanding (SLU) . recent advent of speech-based large language models has opened new avenues for speech understanding .
Approach:	They propose to improve slot-filling task by creating an empirical upper bound for the task . they propose to use a speech-based large language model to integrate speech and text modalities .
Outcome:	The proposed model improves slot filling performance while reducing generalization gaps.

Copied to clipboard

Md Tawkat Islam Khondaker, Julia Harrington, Shady Shehata

Augmenting Compliance-Guaranteed Customer Service Chatbots: Context-Aware Knowledge Expansion with Large Language Models (2025.emnlp-industry)

Challenge:	Recent advances in large language models (LLMs) have significantly transformed medical systems, but their potential within specialized domains such as nursing remains underexplored.
Approach:	They propose a multi-stage data generation pipeline to build the first large scale nursing MCQ dataset to train LLMs on a broad spectrum of nursing topics.
Outcome:	The proposed LLM outperforms general-purpose and medical-specialized LLMs on different benchmarks, underscoring the importance of a specialized Lm for the nursing domain.

Copied to clipboard

Mengze Hong, Chen Jason Zhang, Di Jiang, Yuanqin He

Memory-Efficient Backpropagation for Fine-Tuning LLMs on Resource-Constrained Mobile Devices (2025.emnlp-industry)

Challenge:	Retrieval-based chatbots leverage human-verified Q&A knowledge to deliver accurate, verifiable responses.
Approach:	They propose a similar question generation task for LLM training and inference to enable comprehensive semantic exploration and enhanced alignment with source question-answer relationships.
Outcome:	The proposed methods achieve 92% user satisfaction rate in a deployed chatbot system, reflecting an 18% improvement over the baseline.

Copied to clipboard

Congzheng Song, Xinyu Tang

PrismRAG: Boosting RAG Factuality with Distractor Resilience and Strategized Reasoning (2025.emnlp-industry)

Challenge:	Existing work on memory-efficient on-device fine-tuning of large language models with backpropagation has focused on approximating gradients with zeroth-order optimization (ZO).
Approach:	They propose a memory-efficient implementation of backpropagation on mobile devices that allows flexible trade-offs between memory usage and compute time while converging faster.
Outcome:	The proposed method can fine-tune LLMs with backpropagation using less than 1GB of memory while achieving better performance than the baseline.

Copied to clipboard

Mohammad Kachuee, Teja Gollapudi, Minseok Kim, Yin Huang, Kai Sun, Xiao Yang, Jiaqi Wang, Nirav Shah, Yue Liu, Aaron Colak, Anuj Kumar, Wen-tau Yih, Xin Luna Dong

Benchmarking LLM Faithfulness in RAG with Evolving Leaderboards (2025.emnlp-industry)

Challenge:	Existing methods to improve factuality of large language models (LLMs) rely on human-engineered instructions.
Approach:	They propose a retrieval-augmented generation framework that trains the model with distractor-aware QA pairs mixing gold evidence with subtle distractor passages and instills reasoning-centric habits that make the LLM plan, rationalize, and synthesize without extensive human engineered instructions.
Outcome:	The proposed framework outperforms state-of-the-art solutions across 12 open-book RAG QA benchmarks and is being deployed in production.

Copied to clipboard

Manveer Singh Tamber, Forrest Sheng Bao, Chenyu Xu, Ge Luo, Suleman Kazi, Minseok Bae, Miaoran Li, Ofer Mendelevitch, Renyi Qu, Jimmy Lin

A Multi-Agent Framework for Quantitative Finance : An Application to Portfolio Management Analytics (2025.emnlp-industry)

Challenge:	Large language models (LLMs) excel in various tasks, but often produce hallucinations . retrieved contexts, misrepresent information, or generate outright contradictions .
Approach:	They propose a framework that measures hallucination faithfulness of large language models . they introduce a leaderboard that leverages diverse human-annotated hallucinian examples .
Outcome:	The proposed framework improves hallucination evaluations by leveraging human-annotated examples.

Copied to clipboard

Sayani Kundu, Dushyant Sahoo, Victor Li, Jennifer Rabowsky, Amit Varshney

Group Preference Alignment: Customizing LLM Responses from In-Situ Conversations Only When Needed (2025.emnlp-industry)

Challenge:	Recent advances in Large Language Models (LLMs) have opened up promising new avenues by enhancing reasoning and inference capabilities across diverse data and information sources.
Approach:	They propose a multi-agent framework that facilitates mathematical modeling and data analytics by dynamically generating executable code.
Outcome:	The proposed framework outperforms existing models on portfolio management tasks and provides human-readable explanations for its predictions.

Copied to clipboard

Ishani Mondal, Jack W. Stokes, Sujay Kumar Jauhar, Longqi Yang, Mengting Wan, Xiaofeng Xu, Xia Song, Jordan Lee Boyd-Graber, Jennifer Neville

DASR: Distributed Adaptive Scene Recognition - A Multi-Agent Cloud-Edge Framework for Language-Guided Scene Detection (2025.emnlp-industry)

Challenge:	Existing methods for group-aware adaptation capture divergent preferences from real-world conversation logs into interpretable rubrics.
Approach:	They propose a group-aware personalization framework that captures context-specific preferences and steers LLMs accordingly.
Outcome:	The proposed framework improves group alignment without compromising perfomance on benchmarks.

Copied to clipboard

Can Cui, Yongkang Liu, Seyhan Ucar, Juntong Peng, Ahmadreza Moradipari, Maryam Khabazi, Ziran Wang

Empowering Healthcare Practitioners with Language Models: Structuring Speech Transcripts in Two Real-World Clinical Applications (2025.emnlp-industry)

Challenge:	Current approaches to analyzing driving scenarios rely on massive data collection followed by manual filtering.
Approach:	They propose a cloud-based framework for language-guided scene detection in connected vehicles . the framework leverages cloud- and edge-deployed large language models to identify relevant driving scenarios while optimizing on-vehicle buffer storage.
Outcome:	The proposed framework performs better on complex driving tasks and reduces storage requirements.

Copied to clipboard

Jean-Philippe Corbeil, Asma Ben Abacha, George Michalopoulos, Phillip Swazinna, Miguel Del-Agua, Jerome Tremblay, Akila Jeeson Daniel, Cari Bader, Kevin Cho, Pooja Krishnan, Nathan Bodenstab, Thomas Lin, Wenxuan Teng, Francois Beaulieu, Paul Vozila

Leveraging the Power of Large Language Models in Entity Linking via Adaptive Routing and Targeted Reasoning (2025.emnlp-industry)

Challenge:	Large language models (LLMs) have demonstrated strong performance on clinical natural language processing tasks across multiple medical benchmarks.
Approach:	They propose an agentic pipeline for generating realistic, non-sensitive nurse dictations, enabling structured extraction of clinical observations.
Outcome:	The proposed pipeline generates realistic, non-sensitive nurse dictations, enabling structured extraction of clinical observations.

Copied to clipboard

Yajie Li, Albert Galimov, Mitra Datta Ganapaneni, Pujitha Thejaswi, De Meng, Priyanshu Kumar, Saloni Potdar

Can LLMs Narrate Tabular Data? An Evaluation Framework for Natural Language Representations of Text-to-SQL System Outputs (2025.emnlp-industry)

Challenge:	Entity Linking (EL) relies on large labeled datasets and extensive fine-tuning . lexical ambiguity, knowledge-intensive cases and low-context mentions are some of the challenges.
Approach:	Entity Linking (EL) relies on large annotated datasets and extensive fine-tuning . authors propose a pipeline that integrates candidate generation, context-based scoring, adaptive routing, and selective reasoning .
Outcome:	ARTER outperforms ReFinED and LLM-based pipelines on standard benchmarks . it performs twice as efficiently on 5 out of 6 datasets and doubles the number of tokens compared to pipelines using LLM .

Copied to clipboard

Jyotika Singh, Weiyi Sun, Amit Agarwal, Viji Krishnamurthy, Yassine Benajiba, Sujith Ravi, Dan Roth

Enhancing Foundation Models in Transaction Understanding with LLM-based Sentence Embeddings (2025.emnlp-industry)

Challenge:	Text-to-SQL technology bridges natural language (NL) questions and database querying.
Approach:	They propose a method for evaluating LLM-generated NLRs using Combo-Eval and a dataset for NLR benchmarking.
Outcome:	The proposed method reduces LLM calls by 25-61% and improves performance across scenarios with and without ground truth references.

Copied to clipboard

Xiran Fan, Zhimeng Jiang, Chin-Chia Michael Yeh, Yuzhong Chen, Yingtong Dou, Menghai Pan, Yan Zheng

Agent vs. Agent: Automated Data Generation and Red-Teaming for Custom Agentic Workflows (2025.emnlp-industry)

Challenge:	Existing foundation models for tabular transactional data rely on index-based representations for categorical merchant fields.
Approach:	They propose a framework that uses LLM-generated embeddings as semantic initializations for lightweight transaction models.
Outcome:	The proposed framework improves performance on large transaction datasets.

Copied to clipboard

Ninad Kulkarni, Xian Wu, Siddharth Varia, Dmitriy Bespalov

Auto prompting without training labels: An LLM cascade for product quality assessment in e-commerce catalogs (2025.emnlp-industry)

Challenge:	Existing red-teaming frameworks like AgentHarm use static prompts and hardcoded toolsets .
Approach:	They propose a red-teaming framework that generates adversarial tasks and evaluation functions tailored to arbitrary toolsets and uses iterative prompt refinement with self-reflection to develop more effective attacks.
Outcome:	The proposed approach achieves 162% increase in attack success rate on o4-mini and 86% success on gemini 2.5 Pro.

Copied to clipboard

Soham Satyadharma, Fatemeh Sheikholeslami, Swati Kaul, Aziz Umit Batur, Suleiman A. Khan

Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generation (2025.emnlp-industry)

Challenge:	Our system generates and refines prompts for evaluating attribute quality across tens of thousands of product category–attribute pairs.
Approach:	They propose a free cascade for auto-prompting Large Language Models (LLMs) that generates and refines prompts for evaluating attribute quality across tens of thousands of product category–attribute pairs.
Outcome:	The proposed system improves precision and recall by 8–10% over chain-of-thought prompting while reducing domain expert effort from 5.1 hours to 3 minutes per attribute.

Copied to clipboard

Xujun Peng, Anoop Kumar, Jingyu Wu, Parker Glenn, Daben Liu

Transparent Reference-free Automated Evaluation of Open-Ended User Survey Responses (2025.emnlp-industry)

Challenge:	RAG systems often generate inconsistent outputs for semantically equivalent inputs . this unpredictability undermines the reliability of RAG and poses challenges for adoption in high-stakes or knowledge-sensitive domains such as finance, healthcare, and scientific research.
Approach:	They propose a method that integrates knowledge from specialized models into a single model to improve output consistency.
Outcome:	The proposed model significantly improves output consistency, achieving approximately 47.5% improvement in response similarity over baseline.

Copied to clipboard

Subin An, Yugyeong Ji, Junyoung Kim, Heejin Kook, Yang Lu, Josh Seltzer

Challenge:	Existing methods to evaluate open-ended survey responses are expensive and lack ground-truth reference for comparison.
Approach:	They propose a two-stage evaluation framework specifically designed for human survey responses that uses gibberish filtering to remove nonsensical responses.
Outcome:	The proposed evaluation framework outperforms existing metrics on English and Korean datasets and shows strong correlations with expert assessment.

Datasets and Recipes for Video Temporal Grounding via Reinforcement Learning (2025.emnlp-industry)

Copied to clipboard

Ruizhe Chen, Tianze Luo, Zhiting Fan, Heqing Zou, Zhaopeng Feng, Guiyang Xie, Hansheng Zhang, Zhuochen Wang, Zuozhu Liu, Zhang Huaijian

Visual Data in NLP

Challenge:	Existing methods for video temporal grounding suffer from limited temporal awareness and poor generalization.
Approach:	They propose a two-stage training framework that integrates supervised fine-tuning with reinforcement learning to improve both the accuracy and robustness of VTG models.
Outcome:	The proposed training framework outperforms existing models on multiple benchmarks on open-domain and challenging scenarios.

SEARA: An Automated Approach for Obtaining Optimal Retrievers (2025.emnlp-industry)

Copied to clipboard

Zou Yuheng, Wang Yiran Yiran, Tian Yuzhu, Zhu Min, Yanhua Huang

UniEDU: Toward Unified and Efficient Large Multimodal Models for Educational Tasks (2025.emnlp-industry)

Challenge:	Existing evaluation methods suffer from prohibitive costs or disconnection from domain-specific scenarios.
Approach:	They propose a method which uses subset sampling techniques to obtain robust automated retrieval evaluation at low cost.
Outcome:	The proposed method achieves robust retrieval evaluation by minimal retrieval facts extraction and comprehensive retrieval metrics.

Copied to clipboard

Zhendong Chu, Jian Xie, Shen Wang, Zichao Wang, Qingsong Wen

Truth, Trust, and Trouble: Medical AI on the Edge (2025.emnlp-industry)

Challenge:	Existing research has focused on plain text, while real-world K-12 scenarios often involve multimodal data.
Approach:	They propose a unified language and vision assistant called UniEDU for educational applications . it excels across multiple educational tasks while maintaining strong generalization capabilities . authors propose to use UniEDu for industry-scale deployment .
Outcome:	The proposed model excels across multiple educational tasks while maintaining strong generalization capabilities.

Copied to clipboard

Mohammad Anas Azeez, Rafiq Ali, Ebad Shabbir, Zohaib Hasan Siddiqui, Gautam Siddharth Kashyap, Jiechao Gao, Usman Naseem

Challenge:	Large Language Models (LLMs) are promising for transforming digital health applications . but ensuring they meet industry standards for factual accuracy, usefulness, and safety remains a challenge .
Approach:	They present a framework to assess large language models' accuracy, usefulness, and safety . they assess models' honesty, helpfulness, harmlessness and domain-specific tuning .
Outcome:	The proposed framework assesses models across honesty, helpfulness, and harmlessness . AlpaCare-13B achieves highest accuracy (91.7%) and harmlessity (0.92) .

An Address Intelligence Framework for E-commerce Deliveries (2025.emnlp-industry)

Copied to clipboard

Gokul Swamy, Aman Gulati, Srinivas Virinchi, Anoop Saladi

LLMs on a Budget? Say HOLA (2025.emnlp-industry)

Challenge:	a physical address is an important touchpoint between an e-commerce domain and its customers . incomplete or incorrect addresses can prevent delivery problems and improve the overall customer delivery experience.
Approach:	They propose a language model to assist customers withaddress standardization and a Pareto-ensemble multi-task prediction algorithm that derives critical insights from customer addresses to minimize operational losses.
Outcome:	The proposed system can minimize operational losses in an e-commerce domain.

Copied to clipboard

Zohaib Hasan Siddiqui, Jiechao Gao, Ebad Shabbir, Mohammad Anas Azeez, Rafiq Ali, Gautam Siddharth Kashyap, Usman Naseem

LLM-Based Dialogue Labeling for Multiturn Adaptive RAG (2025.emnlp-industry)

Challenge:	Current solutions such as quantization, pruning, and Retrieval-Augmented Generation (RAG) offer only partial optimizations and often sacrifice accuracy, speed, or generality.
Approach:	They propose an end-to-end optimization framework for efficient LLM deployment . it leverages Hierarchical Speculative Decoding (HSD) for faster inference without quality loss.
Outcome:	HOLA delivers +17.6% EMA on GSM8K, +10.5% MCA on ARC, and reduced latency and memory on edge devices like Jetson Nano.

Copied to clipboard

Zhiyu Chen, Biancen Xie, Sidarth Srinivasan, Manikandarajan Ramanathan, Rajashekar Maragoud, Qun Liu

RAGulator: Lightweight Out-of-Context Detectors for Grounded Text Generation (2025.emnlp-industry)

Challenge:	Retrieval-Augmented Generation (RAG) models integrate large language models with external knowledge retrieval . however, building multi-turn RAG-based chatbots for real-world customer service requires additional complexities.
Approach:	They propose methods to automatically generate labels for adaptive retrieval components using real customer-agent dialogue data.
Outcome:	The proposed method generates labels for components using real customer-agent dialogue data.

Copied to clipboard

Ian Poey, Jiajun Li1, Qishuai Zhong

REIC: RAG-Enhanced Intent Classification at Scale (2025.emnlp-industry)

Challenge:	In enterprise settings, Generative AI has received widespread adoption as a tool to uplift employees' productivity.
Approach:	They develop lightweight models capable of detecting when LLM-generated text deviates from retrieved source documents semantically.
Outcome:	The proposed models outperform open-source alternatives on credit policy and sustainability reports used in the banking industry.

Copied to clipboard

Ziji Zhang, Michael Yang, Zhiyu Chen, Yingying Zhuang, Shu-Ting Pi, Qun Liu, Rajashekar Maragoud, Vy Nguyen, Anurag Beniwal

Mapping Smarter, Not Harder: A Test-Time Reinforcement Learning Agent That Improve Without Labels or Model Updates (2025.emnlp-industry)

Challenge:	Accurate intent classification is critical for efficient routing in customer service . however, as companies expand their product lines, intent classification faces scalability challenges .
Approach:	They propose a retrieval-augmented generation Enhanced Intent Classification approach which leverages retrieval augmented generation to integrate relevant knowledge into a model.
Outcome:	The proposed approach outperforms fine-tuning, zero-shot, and few-shot methods on real-world datasets.

Copied to clipboard

Wen-Kwang Tsao, Yao-Ching Yu, Chien-Ming Huang

Challenge:	a new agent that can improve schema mappings for third-party logs is needed for enterprise intelligence platforms.
Approach:	They propose a reinforcement learning agent that can self-improve without labeled examples or model weight updates.
Outcome:	The proposed method increases mapping accuracy from 56.4% (LLM-only) to 72.73% (RAG) to 93.94% over 100 iterations using GPT-4o.

On Assigning Product and Software Codes to Customer Service Requests with Large Language Models (2025.emnlp-industry)

Copied to clipboard

Sujatha Das Gollapalli, Mouad Hakam, Mingzhe Du, See-Kiong Ng, Mohammed Hamzeh

Reasoning-Enhanced Domain-Adaptive Pretraining of Multimodal Large Language Models for Short Video Content Governance (2025.emnlp-industry)

Challenge:	In a technology company, quality of customer service is a crucial asset.
Approach:	They propose to use Large Language Models to assign product names and software version labels to customer Service Requests (SRs) they frame assignment as multiple-choice question answering task instead of conventional prompts .
Outcome:	The proposed model can identify product names and software versions when they are mentioned with over 90% accuracy while cutting LLM costs by 40-60% on average.

Copied to clipboard

Zixuan Wang, Yu Sun, Hongwei Wang, Baoyu Jing, Xiang Shen, Xin Dong, Zhuolin Hao, Hongyu Xiong, Yang Song

GSID: Generative Semantic Indexing for E-Commerce Product Understanding (2025.emnlp-industry)

Challenge:	Existing approaches to identifying inappropriate content require extensive human-labeled data and lack cross-issue generalization.
Approach:	They propose a reasoning-enhanced multimodal large language model (MLLM) pretraining paradigm for unified inappropriate content detection.
Outcome:	The proposed model improves the MLLM's performance in both zero-shot and supervised fine-tuning settings and shows strong generalization capabilities to emergent, previously unseen issues.

Copied to clipboard

Haiyang Yang, Qinye Xie, Qingheng Zhang, Chen Li Yu, Huike Zou, Chengbao Lian, Shuguang Han, Fei Huang, Jufeng Chen, Bo Zheng

Learning from LLM Agents: In-Context Generative Models for Text Casing in E-Commerce Ads (2025.emnlp-industry)

Challenge:	Structured product information is a major bottleneck for the efficiency of e-commerce platforms.
Approach:	They propose a data-driven approach to generate product structured representations using product metadata.
Outcome:	Extensive experiments show that GSID can generate better product representations on real-world e-commerce platforms.

Copied to clipboard

Yingxue Zhou, Tan Zhu, Tao Zeng, Zigeng Wang, Wei Shen

Auto-Weighted Group Relative Preference Optimization for Multi-Objective Text Generation Tasks (2025.emnlp-industry)

Challenge:	Existing NER-based transformer models are expensive and lack contextual dependencies, making them less reliable when handling unseen or ad-specific terms, e.g., brand names.
Approach:	They propose a two-stage approach to casing correction in e-commerce ad content that leverages Chain-of-Actions to enforce content policies while accurately handling ads-specific terms.
Outcome:	The proposed model outperforms existing NER-based models and achieves near-LLM performance at a fraction of the cost.

Copied to clipboard

Yuki Ichihara, Yuu Jinnai

Cost-Effective E-Commerce Catalog Translation at Scale Ensuring Named Entity Protection (2025.emnlp-industry)

Challenge:	Failing to balance the objectives in advance can lead to overfitting or insufficient learning of each reward function.
Approach:	They propose a method that adjusts reward weights according to learning progress . they evaluate AW-GRPO on advertising text generation problem .
Outcome:	The proposed method outperforms GRPO on advertising text generation tasks . it overfits BLEURT and jReadability at the expense of BLeurT performance .

Copied to clipboard

Asier Gutiérrez-Fandiño, Jorge Yero Salazar, Clement Ruin, Alejandro Quintero-Roba, Shangeetha Ravichandran, Jesus Perez-Martin, Pankaj Adsul, Suruchi Garg, Leonardo Lezcano

InstaJudge: Aligning Judgment Bias of LLM-as-Judge with Humans in Industry Applications (2025.emnlp-industry)

Challenge:	a new translation platform for global e-commerce is delivering high quality, contextually accurate multilingual content.
Approach:	They propose a translation platform for global e-commerce that combines daily batch and real-time API pipelines with optimized T5-based models and a Reference Generator to enforce >99% non-translatable entity preservation.
Outcome:	The proposed translation platform achieves 10–100 cost savings over general-purpose LLMs for EnglishSpanish and EnglishFrench translation.

Copied to clipboard

Myeongjun Erik Jang, Fran Silavong

TelAgentBench: A Multi-faceted Benchmark for Evaluating LLM-based Agents in Telecommunications (2025.emnlp-industry)

Challenge:	Automated evaluation using LLM-as-Judge is a viable alternative to human evaluation, but misalignment of judgment biases between humans and LLMs hinders its use in real-world applications.
Approach:	They propose an LLM-as-Judge library that improves alignments of judgment biases through automatic prompt optimization.
Outcome:	The proposed library outperforms existing LLM-as-Judge libraries by a large margin while being more cost efficient.

Copied to clipboard

Sunwoo Lee, Daseong Jang, Dhammiko Arya, Gyoung-eun Han, Injee Song, SaeRom Kim, Sangjin Kim, Seojin Lee, Seokyoung Hong, Sereimony Sek, Seung-Mo Cho, Sohee Park, Sungbin Yoon, Wonbeom Jang, Eric Davis

Taming the Real-world Complexities in CPT E/M Coding with Large Language Models (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) are becoming powerful agentic systems . generic benchmarks fail to assess realistic, non-English performance .
Approach:	They propose to evaluate five core agentic capabilities: Reasoning, Planning, Action (tool-use), Retrieval-Augmented Generation, and Instruction Following.
Outcome:	The evaluations reveal significant performance disparities between models that employ explicit reasoning and those that do not.

Copied to clipboard

Islam Nassar, Yang Lin, Yuan Jin, Rongxin Zhu, Chang Wei Tan, Zenan Zhai, Nitika Mathur, Thanh Tien Vu, Xu Zhong, Long Duong, Yuan-Fang Li

Classifier-Augmented Generation for Structured Workflow Prediction (2025.emnlp-industry)

Challenge:	Evaluation and Management (E/M) coding is performed by physicians and trained human coders who review clinical encounter notes and electronic health record data to assign appropriate codes.
Approach:	They propose a framework that automates evaluation and management coding tasks using the Current Procedural Terminology (CPT) taxonomy.
Outcome:	The proposed framework achieves an increase in coding accuracy of more than 36% over a commercial CPT E/M coding system and almost 5% over our strongest single-prompt baseline.

Copied to clipboard

Thomas Gschwind, Shramona Chakraborty, Nitin Gupta, Sameep Mehta

Efficient and Versatile Model for Multilingual Information Retrieval of Islamic Text: Development and Deployment in Real-World Scenarios (2025.emnlp-industry)

Challenge:	a new system translates natural language descriptions into executable workflows . configuring stages and their properties is time consuming and requires deep tool knowledge.
Approach:	They propose a system that translates natural language descriptions into executable workflows . it uses a Classifier-Augmented Generation approach that combines utterance decomposition with a classifier and stage-specific prompting to produce accurate stage predictions.
Outcome:	The proposed system outperforms existing models and reduces token usage by 60%.

Copied to clipboard

Vera Pavlova, Mohammed Makhlouf

AutoQual: An LLM Agent for Automated Discovery of Interpretable Features for Review Quality Assessment (2025.emnlp-industry)

Challenge:	Despite recent advances in multilingual information retrieval, a significant gap remains between research efforts and real-world deployment.
Approach:	They propose to use Quranic multilingual corpus to develop an ad-hoc IR system that can satisfy users’ information needs in multiple languages.
Outcome:	The proposed model achieves promising results across diverse retrieval scenarios.

Copied to clipboard

Xiaochong Lan, Jie Feng, Yinxing Liu, null Xinleishi, Yong Li

JSON Whisperer: Efficient JSON Editing with LLMs (2025.emnlp-industry)

Challenge:	Existing methods for assessing review quality are unscalable across domains and fail to adapt to evolving content patterns.
Approach:	They propose an LLM-based agent framework that automates the discovery of interpretable features.
Outcome:	The proposed framework improves on a large-scale online platform with a billion-level user base.

Copied to clipboard

Sarel Duanis, Asnat Greenstein-Messica, Eliya Habba

L4: Mutual Learning Helps Lifelong Language Learning (2025.emnlp-industry)

Challenge:	Large language models can modify JSON documents through natural language commands . current approaches regenerate entire structures for each edit, consuming computational resources .
Approach:	They propose a framework that enables large language models to generate diff patches instead of complete documents.
Outcome:	The proposed framework reduces token usage by 31% while maintaining edit quality within 5% of full regeneration.

Copied to clipboard

Jiyong Li, Dilshod Azizov, Shangsong Liang

TTD-SQL: Tree-Guided Token Decoding for Efficient and Schema-Aware SQL Generation (2025.emnlp-industry)

Challenge:	Existing distillation methods rely on domain-specific teachers, limiting their ability to update in real-time and adapt to dynamic environments.
Approach:	They propose a framework that enables continuous mutual learning from task streams without relying on domain-specific teachers.
Outcome:	The proposed framework reduces catastrophic forgetting while improving performance on various benchmark datasets making it suitable for real-world, dynamic natural language processing (NLP) applications.

Copied to clipboard

Chetan Sharma, Ramasuri Narayanam, Soumyabrata Pal, Kalidas Yeturu, Shiv Kumar Saini, Koyel Mukherjee

Programming Languages in NLP

Challenge:	Large language models (LLMs) have achieved state-of-the-art accuracy on benchmarks like Spider and BIRD, but inference latency due to sequential autoregressive decoding remains a challenge for real-time deployments.
Approach:	a new framework integrates SQL grammar and database schema constraints into the decoding process . tree-Guided Token Decoding (TTD-SQL) precomputes token-level decision trees over SQL keywords, table names, and column identifiers .
Outcome:	a new framework reduces schema hallucinations and inference latency due to autoregressive decoding . tree-Guided Token Decoding achieves 19.96% token-rate speedups .

Spot the BlindSpots: Systematic Identification and Quantification of Fine-Grained LLM Biases in Contact Center Call Summarization (2025.emnlp-industry)

Copied to clipboard

Kawin Mayilvaghanan, Siddhant Gupta, Ayush Kumar

HierDiffuse: Progressive Diffusion for Robust Interest Fusion in CTR Prediction (2025.emnlp-industry)

Challenge:	Abstractive summarization is a core application in contact centers, where Large Language Models generate millions of summaries of call transcripts daily.
Approach:	They propose a framework that uses an LLM as a zero-shot classifier to derive categorical distributions for each bias dimension in a pair of transcripts and its summary.
Outcome:	The proposed framework identifies and quantifies 15 operational bias dimensions and measures them using two metrics: Fidelity Gap and Coverage.

Copied to clipboard

Ziheng Ni, Congcong Liu, Yuying Chen, Zhiwei Fang, Changping Peng, Zhangang Lin, Ching Law, Jingping Shao

TOBUGraph: Knowledge Graph-Based Retrieval for Enhanced LLM Performance Beyond RAG (2025.emnlp-industry)

Challenge:	Existing approaches fuse long-term behavioral profiles and short-term interactions, suffering from representational misalignment and noise in transient signals.
Approach:	They propose a framework that redefines interest fusion as a hierarchical denoising process through diffusion models.
Outcome:	The proposed framework redefines interest fusion as a hierarchical denoising process through diffusion models.

Copied to clipboard

Savini Kashmira, Jayanaka L. Dantanarayana, Joshua Brodsky, Ashish Mahendra, Yiping Kang, Krisztian Flautner, Lingjia Tang, Jason Mars

Thinking with DistilQwen: A Tale of Four Distilled Reasoning and Reward Model Series (2025.emnlp-industry)

Challenge:	Retrieval-Augmented Generation (RAG) relies on query-chunk text-to-text similarity in the embedding space for retrieval, can fail to capture deeper semantic relationships across chunks, is highly sensitive to chunking strategies, and is prone to hallucinations.
Approach:	They propose a graph-based retrieval framework that first constructs the knowledge graph from unstructured data dynamically and automatically.
Outcome:	The proposed framework outperforms multiple RAG implementations in both precision and recall, significantly enhancing user experience through improved retrieval accuracy.

Copied to clipboard

Wenrui Cai, Chengyu Wang, Junbing Yan, Jun Huang, Xiangzhong Fang

Crossing Domains without Labels: Distant Supervision for Term Extraction (2025.emnlp-industry)

Challenge:	In the rapidly evolving landscape of large language models, the need for efficient reasoning models has become increasingly urgent.
Approach:	They extend the Qwen model family by introducing four model series specifically designed for industrial applications.
Outcome:	The proposed models outperform previous models in multiple benchmarks and provide scalable training and inference functionality on the Alibaba Cloud PAI platform.

Copied to clipboard

Elena Senger, Yuri Campbell, Rob Van Der Goot, Barbara Plank

I-SEE: An Instruction-tuned, SOP-Enhanced Quality Evaluator for Product Content (2025.emnlp-industry)

Challenge:	Current state-of-the-art methods require expensive human annotation and struggle with domain transfer, limiting their practical deployment.
Approach:	They propose a benchmark spanning seven diverse domains to evaluate ATE performance . they propose psuedo-labels and post-hoc heuristics to ensure generalizability .
Outcome:	The proposed model outperforms supervised cross-domain encoder models and few-shot learning baselines on the document- and corpus-levels and its GPT-4o teacher on the benchmark.

Copied to clipboard

Aniket Joshi, Cyrus Andre DSouza, Sejal Jain, Jitenkumar Babubhai Rana, Promod Yenigalla

Computational Blueprints: Generating Isomorphic Math Problems with Large Language Models (2025.emnlp-industry)

Challenge:	Existing approaches to content evaluation treat information uniformly without prioritizing based on customer relevance.
Approach:	They propose a framework that combines domain expertise with a single instruction to improve content.
Outcome:	a new framework outperforms existing models in detecting inconsistencies across 20 product categories and 150 product specific features.

Copied to clipboard

Jeong-hoon Kim, Jinwoo Nam, Geunsik Jo

Fin-ExBERT: User Intent based Text Extraction in Financial Context using Graph-Augmented BERT and trainable Plugin (2025.emnlp-industry)

Challenge:	Existing studies on mathematics problem generation focus on data augmentation rather than direct educational deployment.
Approach:	They propose a task to generate structurally consistent variants of source problems . they also propose 'Computational Blueprints for Isomorphic Twins'
Outcome:	The proposed task produces structurally consistent variants of source problems . the authors show that it is superior on generation accuracy and cost-effectiveness at scale .

Copied to clipboard

Soumick Sarker, Abhijit Kumar Rai

DecEx-RAG: Boosting Agentic Retrieval-Augmented Generation with Decision and Execution Optimization via Process Supervision (2025.emnlp-industry)

Challenge:	Financial dialogue transcripts pose a unique challenge for sentence-level information extraction due to their informal structure, domain-specific vocabulary, and variable intent density.
Approach:	They propose a framework for extracting user intent–relevant sentences from financial service calls.
Outcome:	The proposed framework shows strong precision and F1 performance on real-world transcripts . financial transcripts are a challenge due to their informal structure and domain-specific vocabulary .

Copied to clipboard

Yongqi Leng, Yikun Lei, Xikai Liu, Meizhi Zhong, Bojian Xiong, Yurong Zhang, Yan Gao, null Yiwu, Yao Hu, Deyi Xiong

FLOW-BENCH: Towards Conversational Generation of Enterprise Workflows (2025.emnlp-industry)

Challenge:	Recent advances in outcome-supervised reinforcement learning (RL) have shown strong performance, but this approach still suffers from inefficient exploration, sparse reward signals, and ambiguous global reward feedback.
Approach:	They propose a model that models RAG as a Markov Decision Process (MDP) and introduces an efficient pruning strategy to optimize data expansion.
Outcome:	The proposed model outperforms existing methods and achieves an average performance improvement of 6.2% across six datasets.

Copied to clipboard

Evelyn Duesterwald, Siyu Huo, Vatche Isahagian, K. R. Jayaram, Ritesh Kumar, Vinod Muthusamy, Punleuk Oum, Debashish Saha, Gegi Thomas, Praveen Venkateswaran

Format Inertia: A Failure Mechanism of LLMs in Medical Pre-Consultation (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) can be used to convert natural language (NL) instructions into structured business process automation (BPA) process artifacts.
Approach:	They propose to use large language models to convert natural language (NL) instructions into structured business process automation (BPA) process artifacts.
Outcome:	The proposed model can be used to translate NL into Python and convert it into widely adopted business process definition languages.

Copied to clipboard

Seungseop Lim, Gibaeg Kim, Wooseok Han, Jean Seo, Hyunkyung Lee, Jaehyo Yoo, Eunho Yang

Extraction of Information Provision Activity Requirements from EU Acquis (2025.emnlp-industry)

Challenge:	Recent advances in Large Language Models have brought significant improvements to various service domains, including chatbots and medical pre-consultation applications.
Approach:	They propose a method that rebalances the turn-count distribution of training data to mitigate Format Inertia in medical pre-consultation tasks.
Outcome:	The proposed method significantly alleviates Format Inertia in medical pre-consultation tasks.

Copied to clipboard

Jakub Piskorski, Dominik Skotarczak

Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry (2025.emnlp-industry)

Challenge:	Using knowledge-, classical ML-, transformer-, and generative AI-based approaches, we extract structured information from EU acquis documents.
Approach:	They propose a task of Information Provision Activity Requirement Extraction to identify text fragments that introduce an obligation to provide information and the extraction of structured information about the key entities involved.
Outcome:	The proposed task is based on knowledge-, classical ML-, transformer-, and generative AI-based approaches.

Copied to clipboard

Anastasia Zhukova, Jonas Luehrs, Christian Matt, Bela Gipp

From Feedback to Checklists: Grounded Evaluation of AI-Generated Clinical Notes (2025.emnlp-industry)

Challenge:	Recent trends in NLP utilize knowledge graphs to enhance pretrained language models by incorporating additional knowledge from the graph structures to learn domain-specific terminology or relationships between documents that might otherwise be overlooked.
Approach:	They propose to use graph-aware neighborhood contrastive learning methodology SciNCL to enhance pretrained language models by incorporating additional knowledge from graph structures.
Outcome:	The proposed graph-aware neighborhood contrastive learning methodology outperforms a state-of-the-art mE5-large text encoder on the process industry text embedding benchmark while having 3 times fewer parameters.

Copied to clipboard

Karen Zhou, John Michael Giorgi, Pranav Mani, Peng Xu, Davis Liang, Chenhao Tan

Challenge:	Existing automated metrics fail to align with real-world physician preferences.
Approach:	They propose a pipeline that distills real user feedback into structured checklists for note evaluation that are interpretable, grounded in human feedback, and enforceable by LLM-based evaluators.
Outcome:	The proposed checklist outperforms baseline evaluations in coverage, diversity, and predictive power for human ratings.

FlexDoc: Parameterized Sampling for Diverse Multilingual Synthetic Documents for Training Document Understanding Models (2025.emnlp-industry)

Copied to clipboard

Karan Dua, Hitesh Laxmichand Patel, Puneet Mittal, Ranjeet Gupta, Amit Agarwal, Praneet Pabolu, Srikant Panda, Hansa Meghwani, Graham Horwood, Fahad Shah

GEMMAS: Graph-based Evaluation Metrics for Multi Agent Systems (2025.emnlp-industry)

Challenge:	Document understanding models require large, diverse, and well-annotated datasets that can cost millions of dollars to collect and maintain.
Approach:	They propose a scalable synthetic data generation framework that combines Stochastic Schemas and Parameterized Sampling to produce realistic, multilingual semi-structured documents with rich annotations.
Outcome:	Experiments on key information extraction tasks show that the proposed framework improves the absolute F1 score by up to 11% while reducing annotation effort by over 90% compared to traditional hard-template methods.

Copied to clipboard

Jisoo Lee, Raeyoung Chang, Dongwook Kwon, Harmanpreet Singh, Nikhil Verma

Knowledge-Augmented Question Error Correction for Chinese Question Answer System with QuestionRAG (2025.emnlp-industry)

Challenge:	Existing evaluations focus on the correctness of the final output, overlooking inefficient communication and poor coordination contribute to redundant reasoning and higher computational costs.
Approach:	They propose a graph-based evaluation framework that analyzes the internal collaboration process by modeling agent interactions as a directed acyclic graph.
Outcome:	The proposed framework shows that outcome-only metrics are insufficient for evaluating multi-agent performance on GSM8K.

Copied to clipboard

Longpeng Qiu, Ting Li, Shuai Mao, Nan Yang, Xiaohui Yan

SMART: Scalable Multilingual Approach for a Robust TOD System (2025.emnlp-industry)

Challenge:	Large language models struggle with input errors, often failing to interpret user intent or altering the original question’s structure (over-correction).
Approach:	They propose a framework that uses reinforcement learning to address misinterpretation and over-correction by integrating external knowledge with the input.
Outcome:	The proposed framework unlocks the full potential of LLMs for the question correction task.

Copied to clipboard

Karan Malhotra, Arihant Jain, Purav Aggarwal, Anoop Saladi

Think-Search-Patch: A Retrieval-Augmented Reasoning Framework for Repository-Level Code Repair (2025.emnlp-industry)

Challenge:	Existing TOD frameworks face significant challenges in handling unstructured information, providing multilingual support, and engaging proactively.
Approach:	They propose a novel TOD framework that combines traditional pipeline elements with modern agent-based approaches and features a simplified dialogue state, intelligent clarification mechanisms, and a unified natural language generation component that eliminates response redundancy.
Outcome:	The proposed framework outperforms baseline systems across key metrics and integrates in an e-commerce store.

Copied to clipboard

Bojian Xiong, Yikun Lei, Xikai Liu, Shaowei Zhang, Pengyun Zhu, Yan Liu, Yongqi Leng, Ling Shi, Meizhi Zhong, Yurong Zhang, Yan Gao, null Yiwu, Yao Hu, Deyi Xiong

ASR-EC Benchmark: Evaluating Large Language Models on Chinese ASR Error Correction (2025.emnlp-industry)

Challenge:	Large language models suffer from multiple-file coding scenarios with strong inter-file dependencies . experimental results show that large language models exhibit inadequate performance in multi-file scenarios .
Approach:	They propose a retrieval-augmented reasoning framework for repository-level code repair . they use a dataset to generate standardized patches based on the key snippets .
Outcome:	The proposed framework improves retrieval accuracy and repair success on SWE-bench Lite . it surpasses models with larger size in managing extensive code contexts and fixing bugs spanning across multiple files.

Copied to clipboard

Victor Junqiu Wei, Weicheng Wang, Di Jiang, Yuanfeng Song, Lu Wang

Bidirectional Reasoning Supervision for Multilingual Financial Decision Making (2025.emnlp-industry)

Challenge:	Automatic Speech Recognition (ASR) systems have a substantial number of erroneous recognition due to environmental noise, ambiguity, etc.
Approach:	They use a benchmark dataset to analyze ASR errors in the Chinese language . they then apply large language models to correct ASR error correction .
Outcome:	The proposed method is based on a dataset of ASR errors in the Chinese language . it shows prompting is not effective for ASR error correction .

Copied to clipboard

Muhammad Rafsan Kabir, Jawad Ibn Ahad, Robin Krambroeckers, Silvia Ahmed, M M Lutfe Elahi, Nabeel Mohammed, Shafin Rahman

Automotive Document Labeling Using Large Language Models (2025.emnlp-industry)

Challenge:	Large Language Models have been used for sentiment analysis, machine translation, and question answering, but their effectiveness in the multilingual financial domain remains unknown.
Approach:	They propose a fine-tuning approach that integrates positive and negative rationales alongside classification labels.
Outcome:	The proposed approach outperforms existing methods across English, Hindi, Bengali, and Telugu, and is suitable for industry applications.

Copied to clipboard

Dang Van Thin, Cuong Xuan Chu, Christian Graf, Tobias Kaminski, Trung-Kien Tran

Building Data-Driven Occupation Taxonomies: A Bottom-Up Multi-Stage Approach via Semantic Clustering and Multi-Agent Collaboration (2025.emnlp-industry)

Challenge:	Traditionally, mechanics manually browse lengthy documents to locate component information, a process that is time-consuming and error-prone.
Approach:	They propose to use large language models to enrich and unify a component database and use hybrid search to select the most relevant component for a document.
Outcome:	The proposed method outperforms baselines based on an expert-annotated dataset and significantly reduces the search space and improves retrieval efficiency.

Copied to clipboard

Nan Li, Bo Kang, Tijl De Bie

Text Clustering

Challenge:	Existing methods for creating robust occupation taxonomies are slow and expensive . a robust taxonomy is critical for job recommendation and labor market intelligence applications .
Approach:	They propose a framework that automates creation of occupation taxonomies from job postings . they use global semantic clustering to distill core occupations, then a reflection-based multi-agent system to iteratively build a coherent hierarchy.
Outcome:	The proposed framework produces taxonomies that capture unique regional characteristics.

AutoPenBench: A Vulnerability Testing Benchmark for Generative Agents (2025.emnlp-industry)

Copied to clipboard

Luca Gioacchini, Alexander Delsanto, Idilio Drago, Marco Mellia, Giuseppe Siracusano, Roberto Bifulco

Challenge:	LLM agents are promising for vulnerability testing, but lack benchmarks to evaluate and compare them.
Approach:	They propose an open-source benchmark for the evaluation of vulnerability testing agents that includes 33 tasks ranging from introductory exercises to actual vulnerable systems.
Outcome:	The proposed benchmark includes 33 tasks ranging from introductory exercises to actual vulnerable systems.

Enabling Self-Improving Agents to Learn at Test Time With Human-In-The-Loop Guidance (2025.emnlp-industry)

Copied to clipboard

Yufei He, Ruoyu Li, Alex Chen, Yue Liu, Yulin Chen, Yuan Sui, Cheng Chen, Yi Zhu, Luca Luo, Frank Yang, Bryan Hooi

Challenge:	Existing large language model (LLM) agents are unable to adapt to changing domain knowledge and rules.
Approach:	They propose an LLM agent framework that continuously learns updated domain knowledge at test time.
Outcome:	The proposed agent improves on a customer due diligence name screening task on . the agent learns updated domain knowledge at test time.

Encouraging Good Processes Without the Need for Good Answers: Reinforcement Learning for LLM Agent Planning (2025.emnlp-industry)

Copied to clipboard

Zhiwei Li, Yong Hu, Wenqing Wang

Challenge:	Currently, the dominant end-to-end reinforcement learning paradigm for agents in Large Language Models (LLMs) employs multi-objective optimization that jointly trains both planning and answer summarization capabilities.
Approach:	They propose a framework that decouples the training process to enable a focused, single-objective optimization of the planning module.
Outcome:	The proposed framework achieves an 8%–12% improvement in planning performance compared to end-to-end baselines.

Experience Report: Implementing Machine Translation in a Regulated Industry (2025.emnlp-industry)

Copied to clipboard

Marco Zocca, Per Fallgren, David Buffoni

Machine Translation

Challenge:	a global medical technology company has invested substantial resources in translating content into the various languages required across their global markets.
Approach:	They propose to use human-in-the-loop validation to evaluate machine translation systems in a medical technology company.
Outcome:	The proposed method dominates reviewer preference across all languages and tones of interest, the authors show . the "Gold" control ranks poorly in one language and the lower ranks have high variance.

Multi-Task Pre-Finetuning of Lightweight Transformer Encoders for Text Classification and NER (2025.emnlp-industry)

Copied to clipboard

Junyi Zhu, Savas Ozkan, Andrea Maracani, Sinan Mutlu, Cho Jung Min, Mete Ozay

Scaling Down, Serving Fast: Compressing and Deploying Efficient LLMs for Recommendation Systems (2025.emnlp-industry)

Challenge:	nave multitask pre-finetuning introduces conflicting optimization signals that degrade overall performance.
Approach:	They propose a framework that enables a single shared encoder backbone with modular adapters.
Outcome:	The proposed framework achieves comparable performance to individual pre-finetuning while meeting practical deployment constraint.

Copied to clipboard

Kayhan Behdin, Ata Fatahibaarzi, Qingquan Song, Yun Dai, Aman Gupta, Zhipeng Wang, Hejian Sang, Shao Tang, Gregory Dexter, Sirou Zhu, Siyu Zhu, Tejas Dharamsi, Vignesh Kothapalli, Zhoutong Fu, Yihan Cao, Pin-Lun Hsu, Fedor Borisyuk, Natesh S. Pillai, Luke Simon, Rahul Mazumder

Group, Embed and Reason: A Hybrid LLM and Embedding Framework for Semantic Attribute Alignment (2025.emnlp-industry)

Challenge:	Large language models (LLMs) have demonstrated remarkable performance across a wide range of industrial applications.
Approach:	They propose two techniques for training and deploying small language models that deliver high performance for a variety of industry use cases.
Outcome:	The proposed techniques retain much of the quality of larger models while reducing training/serving costs and latency.

Copied to clipboard

Shramona Chakraborty, Shashank Mujumdar, Nitin Gupta, Sameep Mehta, Ronen Kat, Itay Etelis, Mohamed Mahameed, Itai Guez, Rachel Tzoref-Brill

Representation Learning

Challenge:	a framework to align attributes that refer to the same concept but differ across schemas is challenging in schema only settings where no instance data is available due to ambiguous names, inconsistent descriptions, and domain-specific terminologies.
Approach:	They propose a framework that combines contextual reasoning and embedding-based similarity to address token limitations and hallucinations.
Outcome:	The proposed framework scales to large schemas and shows strong performance on healthcare schemas.

STREAQ: Selective Tiered Routing for Effective and Affordable Contact Center Quality Assurance (2025.emnlp-industry)

Copied to clipboard

Prajwal Sood, Rajdeep Agrawal, Mayank Sati, Digvijay Anil Ingle, Cijo George

Divide, Link, and Conquer: Recall-oriented Schema Linking for NL-to-SQL via Question Decomposition (2025.emnlp-industry)

Challenge:	Traditional manual QA cannot scale to growing volumes, while fully automated evaluation using large language models presents a cost-performance trade-off.
Approach:	They propose a two-tier selective routing framework to intelligently route queries between cost-efficient and high-capability models.
Outcome:	The proposed model reduces daily costs by 48% while preserving critical performance.

Copied to clipboard

Kiran Pradeep, Kirushikesh Db, Nishtha Madaan, Sameep Mehta, Pushpak Bhattacharyya

Programming Languages in NLP

Challenge:	Open-source LLMs often depend on large proprietary models, which introduce serious privacy concerns.
Approach:	They propose a plug-and-play framework that improves SQL generation for smaller LLMs . they propose to apply question decomposition at the schema linking stage rather than during SQL generation .
Outcome:	The proposed framework improves schema linking recall by 25.1% and execution accuracy by 8.2% on the BIRD benchmark.

Declarative Techniques for NL Queries over Heterogeneous Data (2025.emnlp-industry)

Copied to clipboard

Elham Khabiri, Jeffrey O. Kephart, Fenno F. Heath Iii, Srideepika Jayaraman, Yingjie Li, Fateh A. Tipu, Dhruv Shah, Achille Fokoue, Anu Bhamidipaty

Challenge:	In many industrial settings, users wish to ask questions in natural language . however, these applications do not cope with data source heterogeneity that typifies such environments.
Approach:	They propose a declarative approach to handling data heterogeneity in industrial settings . they simulate the heterogenity of industrial environments by adding two extensions of the popular Spider benchmark dataset .
Outcome:	The proposed approach copes with data source heterogeneity better than state-of-the-art systems.

Taxonomy of Comprehensive Safety for Clinical Agents (2025.emnlp-industry)

Copied to clipboard

Jean Seo, Hyunkyung Lee, Gibaeg Kim, Wooseok Han, Jaehyo Yoo, Seungseop Lim, Kihun Shin, Eunho Yang

Challenge:	Existing methods for ensuring safety in clinical chatbot applications are not suitable for clinical applications.
Approach:	They propose a fine-grained taxonomy that integrates safety filtering and tool selection into a single user intent classification step.
Outcome:	The proposed taxonomy integrates safety filtering and tool selection into a single user intent classification step.

Dr. Copilot: A Multi-Agent Prompt Optimized Assistant for Improving Patient-Doctor Communication in Romanian (2025.emnlp-industry)

Copied to clipboard

Andrei Niculae, Adrian Cosma, Cosmin Dumitrache, Emilian Radoi

Data-Efficient Automatic Prompt Optimization for Memory-Enhanced Conversational Agents (2025.emnlp-industry)

Challenge:	Text-based telemedicine is increasingly common, but quality of medical advice is judged more on how well it is communicated than on its clinical accuracy.
Approach:	They propose a multi-agent large language model system that evaluates and enhances the presentation quality of doctor-patient interactions.
Outcome:	The proposed system evaluates and improves the presentation quality of Romanian-speaking doctors' written responses.

Copied to clipboard

Ervine Zheng, Yikuan Li, Geoffrey Jay Tso, Jilong Kuang

CLARITY: Clinical Assistant for Routing, Inference, and Triage (2025.emnlp-industry)

Challenge:	Automatic prompt optimization (APO) uses algorithms to optimize prompts for LLMs . but application to memory-enhanced conversational agents presents unique challenges .
Approach:	They propose a framework for automatic prompt optimization for memory-enhanced conversational agents . they leverage LLMs to holistically optimize the prompts of all agents based on memory writing, reading, and response generation .
Outcome:	The proposed framework is applied to memory-enhanced conversational agents . it provides a holistic quality score for responses and performs error attribution .

Copied to clipboard

Vladimir Shaposhnikov, Alexandr Nesterov, Ilia Kopanichuk, Ivan Bakulin, Zhelvakov Egor, Ruslan Abramov, Tsapieva Ekaterina Olegovna, Iaroslav Radionovich Bespalov, Dmitry V. Dylov, Ivan Oseledets

Challenge:	Medical dialogue systems are still flawed for real-world adoption in healthcare.
Approach:	They propose to integrate CLARITY (Clinical Assistant for Routing, Inference and Triage) it combines a Finite State Machine (FSM) and collaborative agents that employ Large Language Model (LLM) they report that it surpasses human-level performance in terms of first-attempt routing precision .
Outcome:	The proposed platform surpasses human-level performance in terms of first-attempt routing precision.

HalluDetect: Detecting, Mitigating, and Benchmarking Hallucinations in Conversational Systems in the Legal Domain (2025.emnlp-industry)

Copied to clipboard

Spandan Anaokar, Shrey Ganatra, Swapnil Bhattacharyya, Harshvivek Kashid, Shruthi N Nair, Reshma Sekhar, Siddharth Manohar, Rahul Hemrajani, Pushpak Bhattacharyya

How Accurate Are LLMs at Multi-Question Answering on Conversational Transcripts? (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) are widely used in industry but still produce hallucinations, limiting their reliability in critical applications.
Approach:	They propose to reduce hallucinations in consumer grievance chatbots by reducing their token accuracy by 0.4159 per turn.
Outcome:	The proposed system achieves an F1 score of 68.92% outperforming baseline detectors by 22.47% while maintaining the highest token accuracy.

Copied to clipboard

Xiliang Zhu, Shi Zong, David Rossouw

AI Knowledge Assist: An Automated Approach for the Creation of Knowledge Bases for Conversational AI Agents (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) are used for question answering over long contexts . high computational costs and latency hinder the process .
Approach:	They explore the capabilities of Large Language Models to answer multiple questions based on the same conversational context.
Outcome:	The proposed models outperform proprietary and public models in question answering . their results show that they can be cost-effective and transparent .

Copied to clipboard

Md Tahmid Rahman Laskar, Julien Bouvier Tremblay, Xue-Yong Fu, Cheng Chen, Shashi Bhushan Tn

DACIP-RC: Domain Adaptive Continual Instruction Pre-Training via Reading Comprehension on Business Conversations (2025.emnlp-industry)

Challenge:	Existing knowledge base is time-consuming and deters the adoption of conversational AI systems in contact centers.
Approach:	They propose a system that extracts knowledge in the form of question-answer (QA) pairs from historical customeragent conversations to automatically build a knowledge base.
Outcome:	The proposed system outperforms larger closed-source LLMs on internal data and achieves above 90% accuracy in answering informationseeking questions.

Copied to clipboard

Elena Khasanova, Harsh Saini, Md Tahmid Rahman Laskar, Xue-Yong Fu, Cheng Chen, Shashi Bhushan Tn

Analysis of Automated Document Relevance Annotation for Information Retrieval in Oil and Gas Industry (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) have been used in real-world industrial scenarios for various natural language processing tasks, but their high inference cost makes their deployment impractical, necessitating the use of smaller models.
Approach:	They propose a continual pre-training technique that generates diverse task instructions and responses via reading comprehension on conversation transcripts, enabling better instruction generalization.
Outcome:	The proposed technique improves small LLMs’ domain adaptability for business conversational tasks, compared with traditional methods that rely on next-token prediction.

Copied to clipboard

João Vitor Mariano Correia, Murilo Missano Bell, João Vitor Robiatti Amorim, Jonas Queiroz, Daniel Pedronette, Ivan Rizzo Guilherme, Felipe Lima de Oliveira

Mind the Query: A Benchmark Dataset towards Text2Cypher Task (2025.emnlp-industry)

Challenge:	Lack of high-quality test collections challenges Information Retrieval (IR) in specialized domains.
Approach:	They compare supervised classifiers against zero-shot Large Language Models for automated relevance annotation in the oil and gas industry using human expert judgments as a benchmark.
Outcome:	The proposed classifier outperforms LLMs in the oil and gas industry using human expert judgments.

Copied to clipboard

Vashu Chauhan, Shobhit Raj, Shashank Mujumdar, Avirup Saha, Anannay Jain

Structured Data in NLP

Challenge:	Graph databases store data in nodes and relationships, enabling more natural modeling of complex, interconnected data.
Approach:	They present a high-quality dataset for the Text2Cypher task . it is enabling the translation of natural language (NL) questions into executable Cypher queries over graph databases.
Outcome:	The proposed dataset includes 27,529 NL queries and corresponding Cyphers spanning across 11 real-world graph datasets.

Deploying Tiny LVLM Judges for Real-World Evaluation of Chart Models: Lessons Learned and Best Practices (2025.emnlp-industry)

Copied to clipboard

Md Tahmid Rahman Laskar, Mohammed Saidul Islam, Ridwan Mahbub, Mizanur Rahman, Amran Bhuiyan, Israt Jahan, Mir Tafseer Nayeem, Shafiq Joty, Enamul Hoque, Jimmy Huang

Agent-in-the-Loop: A Data Flywheel for Continuous Improvement in LLM-based Customer Support (2025.emnlp-industry)

Challenge:	Large Vision-Language Models (LVLMs) with only 7B parameters perform poorly as judges in resource-constrained settings.
Approach:	They propose two approaches to ensure costefficient evaluation by combining multiple criteria into a single query and domainadaptive transfer learning to create a 2Bparameter VLM on a chart dataset.
Outcome:	The proposed model can effectively transfer knowledge from one dataset to another to make it a more specialized model.

Copied to clipboard

Cen Zhao, Tiantian Zhang, Hanchen Su, Yufeng Zhang, Shaowei Su, Mingzhi Xu, Yu Liu, Wei Han, Jeremy Werner, Claire Na Cheng, Yashar Mehdad

Challenge:	Existing offline approaches to improve an LLM-based customer support system rely on batch annotations.
Approach:	They propose an agent-in-the-loop framework that integrates four key types of annotations directly into live customer operations: (1) pairwise response preferences, (2) agent adoption and rationales, (3) knowledge relevance checks, and (4) identification of missing knowledge.
Outcome:	The proposed framework reduces retraining cycles from months to weeks by integrating four key types of annotations directly into live customer operations.

Beyond Pointwise Scores: Decomposed Criteria-Based Evaluation of LLM Responses (2025.emnlp-industry)

Copied to clipboard

Fangyi Yu, Nabeel Seedat, Drahomira Herrmannova, Frank Schilder, Jonathan Richard Schwarz

Scalable and Cost Effective High-Cardinality Classification with LLMs via Multi-View Label Representations and Retrieval Augmentation (2025.emnlp-industry)

Challenge:	DeCE is model-agnostic and domain-general, requiring no predefined taxonomies or handcrafted rubrics.
Approach:	They propose a decomposed LLM evaluation framework that separates accuracy and recall from accuracy and relevance.
Outcome:	The proposed framework achieves stronger correlation with expert judgments than traditional metrics and pointwise LLM scoring.

Copied to clipboard

Anup Pattnaik, Sasanka Vutla, Hamvir Dev, Jeevesh Nandan, Cijo George

How to Fine-Tune Safely on a Budget: Model Adaptation Using Minimal Resources (2025.emnlp-industry)

Challenge:	Existing methods for labeling contact center interactions show significant inconsistencies and sensitivity to label ordering.
Approach:	They propose a two-step retrieval-augmented classification framework enhanced with a multi-view representation of labels.
Outcome:	The proposed method significantly improves accuracy and consistency over baseline methods.

Copied to clipboard

Anh C. Pham, Mihir Thalanki, Michael Sun, Aditya Chaloo, Ankita Gupta, Tian Xia, Aditya Mate, Ehi Nosakhare, Soundararajan Srinivasan

Zero-knowledge LLM hallucination detection and mitigation through fine-grained cross-model consistency (2025.emnlp-industry)

Challenge:	Existing methods for fine-tuning safety examples are underdeveloped.
Approach:	They hypothesize that the effectiveness of a safety example is governed by its instruction-response behavior and its semantic diversity across harm categories.
Outcome:	The proposed method reduces harmfulness by up to 41% while adding only 0.05% more data to the fine-tuning set.

Copied to clipboard

Aman Goel, Daniel Schwartz, Yanjun Qi

Incremental Summarization for Customer Support via Progressive Note-Taking and Agent Feedback (2025.emnlp-industry)

Challenge:	Existing methods for hallucination management fail to integrate both detection and mitigation without external knowledge sources.
Approach:	They propose a black-box framework that leverages fine-grained cross-model consistency to detect and mitigate hallucinations in LLM outputs without external knowledge sources.
Outcome:	The proposed framework improves hallucination detection scores by 6-39% on a FELM dataset . it achieves 9 percentage points improvement in answer accuracy on the GPQA-diamond dataset compared to existing approaches .

Copied to clipboard

Yisha Wu, Cen Zhao, Yuanpei Cao, Xiaoqing Xu, Yashar Mehdad, Mindy Ji, Claire Na Cheng

LLMInit: A Free Lunch from Large Language Models for Selective Initialization of Recommendation (2025.emnlp-industry)

Challenge:	Using a mixtral-87B model, we reduce the time it takes to generate concise notes during conversations, and reduce the amount of time spent on manual review.
Approach:	They propose a system that combines a Mixtral-87B model for continuous note generation with a DeBERTa-based classifier to filter trivial content.
Outcome:	The proposed system achieved a 3% reduction in case handling time compared to bulk summarization, and high agent satisfaction ratings from surveys.

Copied to clipboard

Weizhi Zhang, Liangwei Yang, Wooseong Yang, Henry Peng Zou, Yuqing Liu, Ke Xu, Sourav Medya, Philip S. Yu

LLM Agents Implement an NLG System from Scratch: Building Interpretable Rule-Based RDF-to-Text Generators (2025.emnlp-industry)

Challenge:	Existing algorithms for collaborative filtering are limited by their computational demands and latency.
Approach:	They propose a framework to integrate pre-trained LLM embeddings into CF models through selective initialization strategies.
Outcome:	The proposed framework improves recommendation performance while maintaining low computational costs.

Copied to clipboard

Mateusz Lango, Ondrej Dusek

Leveraging LLMs to Streamline the Review of Public Funding Applications (2025.emnlp-industry)

Challenge:	Existing neural approaches to generate RDF-to-text are limited in their implementation.
Approach:	They propose a framework where the model is “trained” through collaborative interactions among multiple LLM agents rather than traditional backpropagation.
Outcome:	The proposed framework reduces hallucinations and fluency penalties on the WebNLG and OpenDialKG datasets.

Copied to clipboard

João DS Marques, Andre Vicente Duarte, André Mendes Marques de Carvalho, Gil Rocha, Bruno Martins, Arlindo L. Oliveira

AMAS: Adaptively Determining Communication Topology for LLM-based Multi-agent System (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) have been used to improve evaluation processes, but there are concerns over reliability and potential misapplications.
Approach:	They propose to deploy AI-assisted evaluation in two government initiatives . they found that the solution increased reviewer productivity by 20.1% .
Outcome:	The proposed solution reduced reviewer productivity by 20.1% while keeping a negligible false-positive rate . the proposed solution reduces evaluation time by more than 2 months .

Copied to clipboard

Hui Yi Leong, Yuheng Li, Yuqing Wu, Wenwen Ouyang, Wei Zhu, Jiechao Gao, Wei Han

Structured Data in NLP

Challenge:	Large language models (LLMs) have revolutionized natural language processing, but their practical implementation as autonomous multi-agent systems remains fraught with unresolved challenges.
Approach:	They propose a dynamic graph selector that redefines LLM-based MAS by exploiting the intrinsic properties of individual inputs to intelligently direct query trajectories.
Outcome:	The proposed framework exceeds state-of-the-art approaches in question answering, mathematical deduction, and code generation benchmarks.

ColMate: Contrastive Late Interaction and Masked Text for Multimodal Document Retrieval (2025.emnlp-industry)

Copied to clipboard

Ahmed Masry, Megh Thakkar, Patrice Bechard, Sathwik Tejaswi Madhusudhan, Rabiul Awal, Shambhavi Mishra, Akshay Kalkunte Suresh, Srivatsava Daruru, Enamul Hoque, Spandana Gella, Torsten Scholak, Sai Rajeswar

Confidence-Aware Reasoning: Optimizing Self-Guided Thinking Trajectories in Large Reasoning Models (2025.emnlp-industry)

Challenge:	Existing methods for multimodal document retrieval often replicate techniques developed for text-only retrieval.
Approach:	They propose a document retrieval model that bridges the gap between multimodal representation learning and document retrievals by providing external knowledge as context.
Outcome:	The proposed model achieves 3.61% improvement over existing retrieval models on the ViDoRe V2 benchmark, showing stronger generalization to out-of-domain benchmarks.

Copied to clipboard

Jiaxin Zhang

Multi-Value-Product Retrieval-Augmented Generation for Industrial Product Attribute Value Identification (2025.emnlp-industry)

Challenge:	Chain-of-thought enables large reasoning models to reason through multi-step problems but often leads to unnecessarily long or redundant reasoning traces, a phenomenon known as overthinking.
Approach:	They propose an inference-time framework that selectively prunes low-utility reasoning blocks and halts early when sufficient confidence has been achieved.
Outcome:	The proposed framework improves answer accuracy by up to +13.3% while reducing average reasoning length by 40%–50%.

Copied to clipboard

Huike Zou, Haiyang Yang, Yindu Su, Chen Li Yu, Qinye Xie, Chengbao Lian, Qingheng Zhang, Shuguang Han, Fei Huang, Jufeng Chen

AttributeForge: An Agentic LLM Framework for Automated Product Schema Modeling (2025.emnlp-industry)

Challenge:	Existing methods for product attribute value identification suffer from cascading errors and lack of generalization capability.
Approach:	They propose a multi-level retrieval scheme that uses products and attribute values as distinct hierarchical levels in PAVI domain.
Outcome:	The proposed method performs better than the state-of-the-art methods on a real-world industrial dataset.

Copied to clipboard

Yunhan Huang, Klevis Ramo, Andrea Iovine, Melvin Monteiro, Sedat Gokalp, Arjun Bakshi, Hasan Turalic, Arsh Kumar, Jona Neumeier, Ripley Yates, Rejaul Monir, Simon Hartmann, Tushar Manglik, Mohamed Yakout

VestaBench: An Embodied Benchmark for Safe Long-Horizon Planning Under Multi-Constraint and Adversarial Settings (2025.emnlp-industry)

Challenge:	e-commerce platforms are producing only tens of attributes per month for schema modeling . authors present a framework to automate end-to-end product schema modeling using Large Language Models .
Approach:	They introduce a framework to automate end-to-end product schema modeling using Large Language Models.
Outcome:	The proposed framework achieves an 88 increase in modeling throughput while delivering superior quality.

Copied to clipboard

Tanmana Sadhu, Yanan Chen, Ali Pesaranghader

Advancing E-commerce Merchants Telemarketing with Synthetic Data-Driven LLMs (2025.emnlp-industry)

Challenge:	Existing safety benchmarks do not represent a diverse range of multi-constraint tasks that require long-horizon planning with a focus on safety.
Approach:	They propose a benchmark to assess the safety of embodied AI agents under multiple constraints.
Outcome:	The proposed benchmarks show that LLMs perform poorly against their tasks . they also suffer significantly compromised safety outcomes .

Copied to clipboard

Qi Gou, Zehua Xia, Li Juan, Qingyang Zhao, Wenjing Yang

Depression Detection on Social Media with Large Language Models (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) are proving broadly applicable across diverse industries, including e-commerce.
Approach:	They propose a hybrid data synthesis framework that unifies the input schema with profile and strategy designed by top sales and extracts them via a Multi-task paradigm.
Outcome:	The proposed model reaches the performance level of the top 25% of human sales in terms of the final marketing results.

Copied to clipboard

Xiaochong Lan, Zhiguang Han, Yiming Cheng, Li Sheng, Jie Feng, Chen Gao, Yong Li

BullyBench: Youth & Experts-in-the-loop Framework for Intrinsic and Extrinsic Cyberbullying NLP Benchmarking (2025.emnlp-industry)

Challenge:	Existing methods for analyzing social media data lack a systematic integration of medical knowledge, causing a critical treatment gap.
Approach:	They propose a framework that leverages Large Language Models to integrate medical knowledge into social media data.
Outcome:	The proposed framework can be used to distinguish depression from transient mood changes.

Copied to clipboard

Kanishk Verma, Sri Balaaji, Joachim Wagner, Arefeh Kazemi, Darragh Mccashin, Isobel Walsh@dcu, Sayani Basak, Sinan Asci, Yelena Cherkasova, Alexandros Poulis, James Ohiggins Norman, Rebecca Umbach Umbach, Tijana Milosevic, Brian Davis

Text Classification

Challenge:	Existing youth-focused CB datasets lack conversational realism and ethical youth involvement with little or no evaluation of their social plausibility.
Approach:	They propose a youth-in-the-loop dataset “BullyBench” that incorporates a structured intrinsic quality evaluation with experts-in the-looop (social scientists, psychologists, and content moderators) they perform extrinsic baseline evaluation by benchmarking encoder- and decoder-only language models for multi-class CB role classification.
Outcome:	The proposed dataset is evaluated by a team of social scientists, psychologists, and content moderators to assess its quality, relevance, and coherence.

Tagging-Augmented Generation: Assisting Language Models in Finding Intricate Knowledge In Long Contexts (2025.emnlp-industry)

Copied to clipboard

Anwesan Pal, Karen Hovsepian, Tinghao Guo, Mengnan Zhao, Somendra Tripathi, Nikos Kanakaris, George Mihaila, Sumit Nigam

Syntactic Text Processing

Challenge:	Recent studies into effective context lengths of flagship large language models (LLMs) have revealed major limitations in effective question answering (QA) and reasoning over long and complex contexts for even the largest and most impressive cadre of models.
Approach:	They propose a lightweight data augmentation strategy that boosts LLM performance in long-context scenarios without degrading and altering the integrity and composition of retrieved documents.
Outcome:	The proposed strategy boosts performance in long-context scenarios without degrading and altering the integrity and composition of retrieved documents.

DispatchQA: A Benchmark for Small Function Calling Language Models in E-Commerce Applications (2025.emnlp-industry)

Copied to clipboard

Joachim Daiber, Victor Maricato, Ayan Sinha, Andrew Rabinovich

Generalized Embedding Models for Industry 4.0 Applications (2025.emnlp-industry)

Challenge:	DispatchQA is a benchmark to evaluate how well small language models (SLMs) translate openended search queries into executable API calls via explicit function calling.
Approach:	They propose a benchmark to evaluate how well small language models translate openended search queries into executable API calls via explicit function calling.
Outcome:	The proposed benchmark aims to evaluate how well small language models (SLMs) translate openended search queries into executable API calls via explicit function calling.

Copied to clipboard

Christodoulos Constantinides, Shuxin Lin, Dhaval C Patel

ECHO-LLaMA: Efficient Caching for High-Performance LLaMA Training (2025.emnlp-industry)

Challenge:	Using Large Language Models (LLMs) to automate tasks has emerged as the next frontier of innovation.
Approach:	They propose a model that generalizes to queries involving similar assets and retrieves relevant items from natural language tasks.
Outcome:	The proposed model can be used to generalize to queries involving similar assets, such as identifying sensors relevant to an asset’s failure mode.

Copied to clipboard

Maryam Dialameh, Rezaul Karim, Hossein Rajabzadeh, Omar Mohamed Awad, Boxing Chen, Hyock Ju Kwon, Walid Ahmed, Yang Liu

Generating Spatial Knowledge Graphs from Automotive Diagrams for Question Answering (2025.emnlp-industry)

Challenge:	ECHO-LLaMA transforms LLa MA models into shared KV caching across certain layers, significantly reducing KV computational complexity while maintaining or improving language performance.
Approach:	They propose an efficient LLaMA architecture that transforms LLama models into shared KV caching across certain layers, reducing computational complexity while maintaining or improving language performance.
Outcome:	ECHO-LLaMA achieves up to 77% higher token-per-second throughput during training, up to 16% higher Model FLOPs Utilization (MFU) and up to 14% lower loss when trained on an equal number of tokens.

Copied to clipboard

Steve Bakos, Chen Xing, Heidar Davoudi, Aijun An, Ron DiCarlantonio

Enhancing Persuasive Dialogue Agents by Synthesizing Cross‐Disciplinary Communication Strategies (2025.emnlp-industry)

Challenge:	Useful answers require obvious landmarks as a reference point . a decomposed pipeline is the most effective strategy for generating a high-quality SKG .
Approach:	They propose to generate a spatial knowledge graph from a vehicle dashboard diagram . they use large vision-language models to generate the graph using a decomposed pipeline .
Outcome:	The proposed method identifies landmarks with 71.3% agreement with human annotators on a new vehicle dataset.

Copied to clipboard

Shinnosuke Nozue, Yuto Nakano, Yotaro Watanabe, Meguru Takasaki, Shoji Moriya, Reina Akama, Jun Suzuki

BIOPSY - Biomarkers In Oncology: Pipeline for Structured Yielding (2025.emnlp-industry)

Challenge:	Current approaches to developing persuasive dialogue agents rely on predefined persuasive strategies that fail to capture the complexity of real-world interactions.
Approach:	They propose a framework for designing persuasive dialogue agents that draws on proven strategies from social psychology, behavioral economics, and communication theory.
Outcome:	The proposed framework demonstrated significant improvement in the persuasion success rate and generalizability of the datasets.

Copied to clipboard

Sanya A. Chetwani, Jaseem Mahmmdla

DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models (2025.emnlp-industry)

Challenge:	Biomarkers are crucial indicators for early cancer detection and prognosis, but extracting biomarkers from clinical texts remains a challenge.
Approach:	They propose a pipeline that integrates a domain-adapted biomarker entity recognition model and a relation extraction model to link biomarkers to their respective mutations.
Outcome:	The proposed pipeline achieves an F1 score of 0.86 for oncology and 0.87 for neuroscience domains on 5,000 clinical texts.

Copied to clipboard

Yi Shen, Jian Zhang, Jieyun Huang, Shuming Shi, Wenjing Zhang, Jiangze Yan, Ning Wang, Kai Wang, Zhaoxiang Liu, Shiguo Lian

pEBR: A Probabilistic Approach to Embedding Based Retrieval (2025.emnlp-industry)

Challenge:	Recent advances in slow-thinking reasoning models have shown exceptional performance in complex reasoning tasks.
Approach:	They propose a framework that enables models to automatically adjust Chain-of-Thought (CoT) length based on problem difficulty.
Outcome:	The proposed framework penalizes inefficiency on simple problems while incentivizing deep reasoning for complex ones.

Copied to clipboard

Han Zhang, Yunjiang Jiang, Mingming Li, Haowei Yuan, Yiming Qiu, Wen-Yun Yang

Finding Diamonds in Conversation Haystacks: A Benchmark for Conversational Data Retrieval (2025.emnlp-industry)

Challenge:	Existing embedding-based retrieval systems rely on heuristic and suboptimal cutoffs for item retrieval.
Approach:	They propose a probabilistic Embedding-Based Retrieval framework that learns a shared semantic representation space for both queries and items.
Outcome:	The proposed framework improves retrieval precision and recall, and ablation studies show it captures the differences between head-to-tail queries.

Copied to clipboard

Yohan Lee, Yongwoo Song, Sangyeop Kim

<SYNTACT>: Structuring Your Natural Language SOPs into Tailored Ambiguity-Resolved Code Templates (2025.emnlp-industry)

Challenge:	Our work identifies unique challenges in conversational data retrieval . large language model-based systems operate through open-ended interactions without predefined specifications.
Approach:	They propose a benchmark to evaluate systems that retrieve conversation data for product insights.
Outcome:	The benchmark provides a reliable standard for measuring conversational data retrieval performance.

Copied to clipboard

Sachin Kumar Giroh, Pushpendu Ghosh, Aryan Jain, Harshal Giridhari Paunikar, Aditi Rastogi, Promod Yenigalla, Anish Nediyanchath

Challenge:	Unstructured and ambiguous Standard Operating Procedures suffer from ambiguity, missing information, and inconsistency, all of which hinder automation.
Approach:	They propose a three-stage LLM framework that transforms unstructured SOPs into a structured plan and an executable code template.
Outcome:	The proposed framework shows an 88.4% accuracy and significant reduction in inconsistency on real-world SOPs and synthetic variants.

Recover-LoRA: Data-Free Accuracy Recovery of Degraded Language Models via Low-Rank Adaptation (2025.emnlp-industry)

Copied to clipboard

Devleena Das, Rajeev Patwari, Ashish Sirasao

ixi-GEN: Efficient Industrial sLLMs through Domain Adaptive Continual Pretraining (2025.emnlp-industry)

Challenge:	Inference optimizations such as quantization, pruning, format and datatype conversion, model export, and serialization can lead to functional degradations in language model task performance.
Approach:	They propose a lightweight and dataset-agnostic method to recover model accuracies from quantization, pruning, format and datatype conversion, model export, and serialization errors.
Outcome:	The proposed method recovers model accuracies by 5-17% on MHA and GQA models.

Copied to clipboard

Seonwu Kim, Yohan Na, Kihun Kim, Hanhee Cho, Geun Lim, Mintae Kim, Seongik Park, Ki Hyun Kim, Youngsub Han, Byoung-Ki Jeon

GRAFT: A Graph-based Flow-aware Agentic Framework for Document-level Machine Translation (2025.emnlp-industry)

Challenge:	Domain Adaptive Continual Pretraining (DACP) is a method to mitigate performance degradation in small LLMs and enhance their effectiveness in target domains.
Approach:	They propose a continual pretraining methodology that optimizes sLLMs within service domains and enhances their effectiveness in target domains.
Outcome:	The proposed model achieves significant gains in target-domain performance while preserving general capabilities, offering a cost-efficient and scalable solution for enterprise-level deployment.

Copied to clipboard

Himanshu Dutta, Sunny Manchanda, Prakhar Bapat, Meva Ram Gurjar, Pushpak Bhattacharyya

Recon, Answer, Verify: Agents in Search of Truth (2025.emnlp-industry)

Challenge:	Existing Document-level machine translation systems struggle to handle discourse-level phenomena such as pronoun resolution, lexical cohesion, and ellipsis.
Approach:	They propose a graph-based document-level machine translation framework that leverages Large Language Models to model translation flow and discourse structure.
Outcome:	The proposed framework outperforms commercial and closed systems in eight languages and six domains.

Copied to clipboard

Satyam Shukla, Himanshu Dutta, Pushpak Bhattacharyya

T-VEC: A Telecom-Specific Vectorization Model with Enhanced Semantic Understanding via Deep Triplet Loss Fine-Tuning (2025.emnlp-industry)

Challenge:	Existing benchmark datasets suffer from leakage or evidence incompleteness, limiting the realism of current evaluations.
Approach:	They propose an agentic framework that iteratively generates and answers sub-questions to verify different aspects of the claim before finally generating the label.
Outcome:	The proposed system outperforms existing methods by 57.5% on Politi-Fact-Only and 3.05% on the widely used HOVER datasets.

Copied to clipboard

Vignesh Ethiraj, Ashwath D, Sidhanth Menon, Divya Vijay, Vidhyakshaya Kannan

PlanGPT-VL: Enhancing Urban Planning with Domain-Specific Vision-Language Models (2025.emnlp-industry)

Challenge:	Generic embedding models struggle to represent telecom-specific semantics . specialized terminology and ambiguous terms often limit their utility in retrieval and downstream tasks.
Approach:	They propose a domain-adapted embedding model fine-tuned from a gte-Qwen2-1.5B-instruct backbone.
Outcome:	The proposed model outperforms MPNet, BGE, Jina and E5 on a custom benchmark . it is open source and has a triplet loss objective .

Copied to clipboard

He Zhu, Junyou Su, Minxin Chen, Wen Wang, Yijie Deng, Guanhua Chen, Wenjia Zhang

IPR: Intelligent Prompt Routing with User-Controlled Quality-Cost Trade-offs (2025.emnlp-industry)

Challenge:	Existing Vision-Language Models (VLMs) fail to analyze planning maps . specialized visual representations of land use zones, transportation networks, and development policies are needed to interpret complex planning maps.
Approach:	They propose a domain-specific VLM tailored for urban planning maps that employs three innovations: PlanAnno-V framework for high-quality VQA data synthesis, Critical Point Thinking (CPT) and PlanBench-V benchmark for systematic evaluation.
Outcome:	The new model outperforms general-purpose VLMs on planning map interpretation tasks.

Copied to clipboard

Aosong Feng, Balasubramaniam Srinivasan, Yun Zhou, Zhichao Xu, Kang Zhou, Sheng Guan, Yueyan Chen, Xian Wu, Ninad Kulkarni, Yi Zhang, Zhengyuan Shen, Dmitriy Bespalov, Soumya Smruti Mishra, Yifei Teng, Darren Yow-Bang Wang, Haibo Ding, Lin Lee Cheong

Low-Resource NLP

Challenge:	Existing systems require users to manually select models or employ rigid routing rules that fail to capture the continuous spectrum of query complexity.
Approach:	They propose a quality-constrained intelligent prompt routing framework that automatically selects optimal models based on predicted response quality and user-specified tolerance levels.
Outcome:	The proposed framework achieves 43.9% cost reduction while maintaining quality parity with strongest model in the Claude family and processes requests with sub-150ms latency.

Semantic Agreement Enables Efficient Open-Ended LLM Cascades (2025.emnlp-industry)

Copied to clipboard

Duncan Soiffer, Steven Kolawole, Virginia Smith

Lost in Pronunciation: Detecting Chinese Offensive Language Disguised by Phonetic Cloaking Replacement (2025.emnlp-industry)

Challenge:	Large language models (LLMs) have enabled impressive progress across a range of language tasks, but they are steep computational cost.
Approach:	They propose a semantic consensus mechanism for reliable deferral by combining model outputs with semantic consensus.
Outcome:	Evaluated from 500M to 70B-parameter models, semantic cascades improve deferral accuracy, match or surpass target-model quality at 40% of the cost and reduce latency by up to 60%.

Copied to clipboard

Haotan Guo, Jianfei He, Jiayuan Ma, Hongbin Na, Zimu Wang, Haiyang Zhang, Qi Chen, Wei Wang, Zijing Shi, Tao Shen, Ling Chen

Distilling Cross-Modal Knowledge into Domain-Specific Retrievers for Enhanced Industrial Document Understanding (2025.emnlp-industry)

Challenge:	Phonetic Cloaking Replacement (PCR) is a problem in content moderation in China.
Approach:	They organize PCR into a four-way surface-form taxonomy and compile PCR-ToxiCN, a dataset of 500 phonetically cloaked offensive posts gathered from the RedNote platform.
Outcome:	The proposed model achieves only an F1-score and zero-shot chain-of-thought prompting pushes performance even lower.

Copied to clipboard

Jinhyeong Lim, Jeongwan Shin, Seeun Lee, Seongdeok Kim, Joungsu Choi, Jongbae Kim, Chun Hwan Jung, Youjin Kang

Don’t Forget the Base Retriever! A Low-Resource Graph-based Retriever for Multi-hop Question Answering (2025.emnlp-industry)

Challenge:	Retrieval-Augmented Generation (RAG) has shown strong performance in open-domain tasks, but its effectiveness in industrial domains is limited by a lack of domain understanding and document structural elements (DSE) such as tables, figures, charts, and formula.
Approach:	They propose a knowledge distillation framework that transfers complementary knowledge from Large Language Models and Vision-Language Models into a compact domain-specific retriever.
Outcome:	The proposed framework outperforms larger baselines while requiring significantly less computational complexity.

Copied to clipboard

Andre Melo, Enting Chen, Pavlos Vougiouklis, Chenxin Diao, Shriram Piramanayagam, Ruofei Lai, Jeff Z. Pan

Structured Data in NLP

Challenge:	Existing GraphRAG approaches to multi-hop question answering rely on expensive LLM calls.
Approach:	They propose a lightweight, low-resource, multi-step graph-based retriever for multi-hop QA that performs multi- step retrieval in a few hundred milliseconds.
Outcome:	The proposed retriever outperforms conventional retrievers on multi-hop QA datasets and shows strong potential as a base retriever within multi-step agentic frameworks.

Beyond Dynamic Quantization: An Efficient Static Hierarchical Mix-precision Framework for Near-Lossless LLM Compression (2025.emnlp-industry)

Copied to clipboard

Yi Zhang, Kai Zhang, Zheyang Li, Wenming Tan, Ye Ren, Jilin Hu

STACKFEED: Structured Textual Actor-Critic Knowledge base editing with FEEDback (2025.emnlp-industry)

Challenge:	Existing methods for dynamic quantization are hardware-unfriendly and often lead to large quantization errors in static scenarios.
Approach:	They propose a Static Hierarchical Mix-precision Quantization method which quantifies both inter-layer and intra-layer sensitivity through unified derivations involving Hessian.
Outcome:	The proposed method achieves 75.58% on zero-shot reasoning tasks while yielding average speedup of 2.86.

Copied to clipboard

Shashank Kirtania, Naman Gupta, Priyanshu Gupta, Sumit Gulwani, Arun Iyer, Suresh Parthasarathy Iyengar, Arjun Radhakrishna, Sriram K. Rajamani, Gustavo Soares

JaCorpTrack: Corporate History Event Extraction for Tracking Organizational Changes (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) often generate incorrect or outdated information, especially in low-resource settings or when dealing with private data.
Approach:	They propose a framework that iteratively refines the knowledge base based on expert feedback . they define a ReACT actor agent on each document to perform structured edits .
Outcome:	The proposed framework improves the quality and performance of the RAG system on low-resource programming problems, modified Python packages, and factual question-answering tasks.

Copied to clipboard

Yuya Sawada, Hiroki Ouchi, Yuichiro Yasui, Hiroki Teranishi, Yuji Matsumoto, Taro Watanabe, Masayuki Ishii

CTR-Guided Generative Query Suggestion in Conversational Search (2025.emnlp-industry)

Challenge:	Existing information extraction systems are not able to accurately capture organizational changes.
Approach:	They propose a task to extract corporate history events related to organizational changes by identifying company names before and after each event, as well as the corresponding date.
Outcome:	The proposed task is designed to identify company names before and after an event, as well as the corresponding date.

Copied to clipboard

Erxue Min, Hsiu-Yuan Huang, Xihong Yang, Min Yang, Xin Jia, Yunfang Wu, Hengyi Cai, Junfeng Wang, Shuaiqiang Wang, Dawei Yin

LATTE: Learning Aligned Transactions and Textual Embeddings for Bank Clients (2025.emnlp-industry)

Challenge:	Generating effective query suggestions requires aligning model outputs with user click preferences.
Approach:	They propose a generative framework that leverages click modeling to denoise implicit feedback and enables reliable preference optimization for improving real-world user engagement.
Outcome:	The proposed framework outperforms strong baselines in CTR, relevance, diversity and diversity.

Copied to clipboard

Egor Fadeev, Dzhambulat Mollaev, Aleksei Shestov, Dima Korolev, Omar Zoloev, Ivan A Kireev, Andrey Savchenko, Maksim Makarenko

RedOne: Revealing Domain-specific LLM Post-Training in Social Networking Services (2025.emnlp-industry)

Challenge:	Large language models (LLMs) are computationally expensive and impractical for real-world pipelines.
Approach:	They propose a contrastive learning framework that aligns raw event embeddings with description-based semantic embedds from frozen LLMs.
Outcome:	The proposed framework outperforms state-of-the-art techniques for learning event sequence representations on real-world financial datasets while remaining deployable in latency-sensitive environments.

Copied to clipboard

Fei Zhao, Chonggang Lu, null Wangyue, Zheyong Xie, Ziyan Liu, Haofu Qian, Jianzhao Huang, Fangcheng Shi, Zijie Meng, Hongcheng Guo, Mingqian He, Xinze Lyu, Zheyu Ye, Weiting Liu, Boyang Wang, Shaosheng Cao

High-Quality Medical Dialogue Synthesis for Improving EMR Generation (2025.emnlp-industry)

Challenge:	Social networking services (SNS) have experienced rapid growth, which has proposed significant challenges for platform content management and interaction quality improvement.
Approach:	They propose a domain-specific LLM to break the performance bottleneck of single-task baselines and establish a comprehensive foundation for social networking services.
Outcome:	The proposed model achieves an average improvement of 14.02% across 8 major tasks and 7.56% in bilingual evaluation benchmark, compared with baseline models.

Copied to clipboard

Chengze Ge, Yu Xu, Qi Shao, Shengping Liu

Z1: Efficient Test-time Scaling with Code (2025.emnlp-industry)

Challenge:	Existing methods for generating EMRs from doctor-patient dialogues produce rigid and repetitive dialogues.
Approach:	They propose a framework that integrates Intent Graph Planning, Dual-Agent Simulation and Rule-Reward Quality Control to generate realistic doctor-patient dialogues.
Outcome:	The proposed framework significantly enhances realism, diversity and downstream EMR quality, reducing physician editing efforts.

Copied to clipboard

Zhaojian Yu, Yinghao Wu, Yilun Zhao, Arman Cohan, Xiao-Ping Zhang

Quality Assessment of Tabular Data using Large Language Models and Code Generation (2025.emnlp-industry)

Challenge:	Large Language Models (LLMs) can achieve enhanced complex problem-solving through test-time computing scaling, but this often entails longer contexts and numerous reasoning token costs.
Approach:	They propose an efficient test-time scaling method that trains LLMs on code-related reasoning trajectories and a novel Shifted Thinking Window to mitigate overthinking overhead.
Outcome:	The proposed method reduces overthinking overhead while maintaining performance.

Copied to clipboard

Ashlesha Akella, Akshar Kaul, Krishnasuri Narayanam, Sameep Mehta

PARSE: LLM Driven Schema Optimization for Reliable Entity Extraction (2025.emnlp-industry)

Challenge:	Data quality is vital for business decisions; poor data quality costs organizations an average of $12.9 million annually.
Approach:	They propose a framework that combines statistical inliner detection with LLM-driven rule and code generation.
Outcome:	The proposed framework produces semantically valid quality rules and validates them with retrieval-augmented generation (RAG) Extensive evaluations on benchmark datasets confirm the effectiveness of the proposed framework.

Copied to clipboard

Anubhav Shrimal, Aryan Jain, Soumyajit Chowdhury, Promod Yenigalla

From Long Videos to Engaging Clips: A Human-Inspired Video Editing Framework with Multimodal Narrative Understanding (2025.emnlp-industry)

Challenge:	Structured information extraction from unstructured text is critical for Software 3.0 systems . current approaches to extract structured information from unstructed text are static contracts .
Approach:	They propose a system that automates JSON schemas for LLM consumption and optimizes them for LRM consumption.
Outcome:	The proposed system improves extraction accuracy and reduces errors by 92% within the first retry and maintaining practical latency.

Copied to clipboard

Xiangfeng Wang, Xiao Li, Yadong Wei, null Songxueyu, Yang Song, null Xiaxiaoqiang, Fangrui Zeng, Zaiyi Chen, null Liuliu, Gu Xu, Tong Xu

Efficiency-Effectiveness Reranking FLOPs for LLM-based Rerankers (2025.emnlp-industry)

Challenge:	Existing methods for video editing rely on textual cues from ASR transcripts and segment selection, often neglecting rich visual context.
Approach:	They propose a human-inspired automatic video editing framework that leverages multimodal narrative understanding to address these limitations.
Outcome:	The proposed framework outperforms existing baselines across general and advertisement-oriented editing tasks.

Copied to clipboard

Zhiyuan Peng, Ting-Ruen Wei, Tingyu Song, Yilun Zhao

GEAR: A Scalable and Interpretable Evaluation Framework for RAG-Based Car Assistant Systems (2025.emnlp-industry)

Challenge:	Existing studies evaluate the efficiency of LLM-based rerankers using proxy metrics such as latency and the number of forward passes.
Approach:	They propose to use a large language model to evaluate the efficiency of LLM-based rerankers . they propose to measure ranking quality and query processing efficiency using an interpretable FLOPs estimator .
Outcome:	The proposed metrics evaluate LLM-based rerankers with different architectures without running any experiments.

Copied to clipboard

Niloufar Beyranvand, Hamidreza Dastmalchi, Aijun An, Heidar Davoudi, Winston Chan, Ron DiCarlantonio

FQ-Eval: Building Evaluation Dataset for User-centered Follow-up Question Generation (2025.emnlp-industry)

Challenge:	Large language models (LLMs) increasingly power car assistants, but evaluating response quality remains a challenge.
Approach:	They propose a framework that uses large language models as evaluators to compare assistant responses against ground-truth counterparts.
Outcome:	The proposed framework compares assistant responses against ground-truth counterparts, assessing coverage, correctness, and other dimensions of answer quality.

Copied to clipboard

Sanghyun Seo, Bumsoo Kang, Dahm Lee, Jaeheon Kim, Joongbo Shin, Eui Soon Kim, Kijeong Jeon

Evaluating AI for Finance: Is AI Credible at Assessing Investment Risk Appetite? (2025.emnlp-industry)

Challenge:	Existing studies focus on enhancing information-seeking or topical relevance, often missing how follow-up questions could satisfy users’ intrinsic needs and conversational goals.
Approach:	They propose a user-centered evaluation dataset for assessing follow-up question generation in chat-LLM services that incorporates realistic chat-llm usage scenarios and five distinct human-aligned criteria.
Outcome:	The proposed model captures human-aligned criteria for the evaluation of various models and services.

Copied to clipboard

Divij Chawla, Ashita Bhutada, Duc Anh Do, Abhinav Raghunathan, Vinod Sp, Cathy Guo, Dar Win Liew, Prannaya Gupta, Rishabh Bhardwaj, Rajat Bhardwaj, Soujanya Poria

CAPSTONE: Composable Attribute‐Prompted Scene Translation for Zero‐Shot Vision–Language Reasoning (2025.emnlp-industry)

Challenge:	Our analysis was conducted on proprietary systems and open-weight models . FINRISKEVAL analyzed 1,720 profiles spanning a broad spectrum of possible risk categories .
Approach:	They evaluated proprietary AI systems and open-weight models to assess investment risk appetite using carefully curated user profiles.
Outcome:	The proposed models exhibit significant variance when user attributes that should not influence risk computation are changed.

Copied to clipboard

Md. Ismail Hossain, Shahriyar Zaman Ridoy, Moshiur Farazi, Nabeel Mohammed, Shafin Rahman

Building Resource-Constrained Language Agents: A Korean Case Study on Chemical Toxicity Information (2025.emnlp-industry)

Challenge:	CAPSTONE transforms visual inputs into structured text prompts that can be interpreted by a frozen Large Language Model (LLM).
Approach:	They propose a plug-and-play framework that transforms off-the-shelf vision models into structured text prompts that can be interpreted by a frozen Large Language Model (LLM).
Outcome:	The proposed framework outperforms fully trained VLMs on the POPE dataset while the 4B model achieves competitive results.

Copied to clipboard

Hojun Cho, Donghu Kim, Soyoung Yang, Chan Lee, Hunjoo Lee, Jaegul Choo

Challenge:	Existing language agents powered by large language models face resource-constrained environments . proprietary models raise concerns in cost and service dependency, while large-scale open-source models require substantial computational resources.
Approach:	They propose a Korean chemical toxicity information agent that reduces token consumption . they propose 'scenario-based dialogue generation' methodology that distills tool-using capabilities from larger models.
Outcome:	The proposed language agent outperforms untuned models and baseline approaches in DB faithfulness and preference.

AutoDSPy: Automating Modular Prompt Design with Reinforcement Learning for Small and Large Language Models (2025.emnlp-industry)

Copied to clipboard

Nafew Azim, Abrar Ur Alam, Hasan Bin Omar, Abdullah Mohammad Muntasir Adnan Jami, Jawad Ibn Ahad, Muhammad Rafsan Kabir, Md. Ismail Hossain, Fuad Rahman, Mohammad Ruhul Amin, Shafin Rahman, Nabeel Mohammed