China

In this paper, we use large language models to generate personalized stories for language learners, using only the vocabulary they know.
The generated texts are specifically written to teach the user new vocabulary by simply reading stories where it appears in context, while at the same time seamlessly reviewing recently learned vocabulary. The generated stories are enjoyable to read and the vocabulary reviewing/learning is optimized by a Spaced Repetition System.
The experiments are conducted in three languages: English, Chinese and Polish, evaluating three story generation methods and three strategies for enforcing lexical constraints. The results show that the generated stories are more grammatical, coherent, and provide better examples of word usage than texts generated by the standard constrained beam search approach.

EMNLP 2025

SRS-Stories: Vocabulary-constrained multilingual story generation for language learning

poster

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community."<br>

*Message from the General Chair, Dirk Hovy*

[**Link to Conference Handbook**](https://drive.google.com/file/d/1johU5QqVVYO4RfH7QcIORr7qrVBdzdwC/view?usp=sharing)





<br>

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

Given the inherent subjectivity of similarity in text, fully unsupervised text clustering is unlikely to produce groupings that work across a variety of use cases. Traditional techniques to guide clustering rely on costly, time-consuming human feedback and/or pre-existing labels. Leveraging recent advancements in LLMs and decoder-only embedding models, we present techniques to effectively control text embeddings with minimal human input: prefix instructions and LLM preprocessing. We evaluate clustering performance for datasets with multiple independent ground-truth labels, or perspectives, and find that these techniques can be used to improve clustering for one perspective or use case, at the cost of a tradeoff in performance for another use case.

Controllable Clustering with LLM-driven Embeddings

Open-ended survey responses provide valuable insights in marketing research, but low-quality responses not only burden researchers with manual filtering but also risk leading to misleading conclusions, underscoring the need for effective evaluation. Existing automatic evaluation methods target LLM-generated text and inadequately assess human-written responses with their distinct characteristics. We propose a two-stage evaluation framework specifically designed for human survey responses. First, gibberish filtering filters out nonsensical responses, then three dimensions—effort, relevance, and completeness—are evaluated using LLM capabilities, grounded in empirical analysis of real-world survey data. Validation on English and Korean datasets shows that our framework outperforms existing metrics and demonstrates high practical applicability for real-world applications across multilingual setting, showing strong correlations with expert assessment.

Transparent Reference-free Automated Evaluation of Open-Ended User Survey Responses

Real-time identification of out-of-context outputs from large language models (LLMs) is crucial for enterprises to safely adopt retrieval augmented generation (RAG) systems. In this work, we develop lightweight models capable of detecting when LLM-generated text deviates from retrieved source documents semantically. We compare their performance against open-source alternatives on data from credit policy and sustainability reports used in the banking industry. The fine-tuned DeBERTa model stands out for its superior performance, speed, and simplicity, as it requires no additional preprocessing or feature engineering. While recent research often prioritises state-of-the-art accuracy through fine-tuned generative LLMs and complex training pipelines, we demonstrate how detection models are deployed efficiently with high speed and minimal resource usage.

RAGulator: Lightweight Out-of-Context Detectors for Grounded Text Generation

Group Relative Policy Optimization (GRPO) is a promising approach to complex, real-world tasks, such as those involving multiple rewards or strict constraints. However, when training GRPO with multiple rewards, the weights of each reward must be decided in advance. Failing to balance the objectives adequately can lead to overfitting or insufficient learning of each reward function. 
To address this problem, we propose Auto-Weighted Group Relative Policy Optimization (AW-GRPO), which adjusts reward weights during training according to the progress of the learning of each objective so far.
We evaluate AW-GRPO on advertising text generation, a real-world problem where the generated text must satisfy multiple objectives, such as quality and diversity, while adhering to the constraints of the media (e.g., maximum number of characters).
Our results show that AW-GRPO successfully balances multiple objectives, improving the overall scores while reducing the constraint violation rate.
We additionally evaluate AW-GRPO using publicly available benchmark problems for reproducibility, in which we observe the same qualitative result that the proposed method outperforms GRPO.

Auto-Weighted Group Relative Preference Optimization for Multi-Objective Text Generation Tasks

We present an enterprise‐grade 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. A linguist‐driven rule engine and explainable evaluation framework (BLEU, COMET, and a custom e-commerce metric) enable continuous quality improvements. Deployed on GPU-accelerated inference servers and CPU-based processing nodes, our system processes millions of listings per day with sub-second latency and achieves 10×–100× cost savings over general-purpose LLMs for English→Spanish and English→French translation, all while version-tracking every update for robust enterprise rollouts.

Cost-Effective E-Commerce Catalog Translation at Scale Ensuring Named Entity Protection

Automated evaluation using LLM-as-Judge offers significant practical benefits for industrial applications. However, the commonly recognized misalignment of judgment biases between humans and LLM-as-Judge hinders its usage in real-world businesses. Although preference-finetuning could be a potential solution, it is often impractical for industrial use-cases due to the scarcity of business-specific data and the infeasibility of applying it to closed models. In this paper, we propose InstaJudge, an LLM-as-Judge library that improves alignments of judgment biases through automatic prompt optimization (APO). Our library not only integrates recent APO methods within a unified framework but also introduces a novel APO approach called distribution-preserving few-shot sampling (DPFS). Experimental results verify demonstrate DPFS significantly outperforms existing LLM-as-Judge libraries, like DeepEval, and APO methods by a large margin, while being more cost efficient.

InstaJudge: Aligning Judgment Bias of LLM-as-Judge with Humans in Industry Applications

ETL (Extract, Transform, Load) tools such as IBM DataStage allow users to visually assemble complex data workflows, but configuring stages and their properties remains time-consuming and requires deep tool knowledge. We propose a system that translates natural language descriptions into executable workflows, automatically predicting both the structure and detailed configuration of the flow. At its core lies a Classifier-Augmented Generation (CAG) approach that combines utterance decomposition with a classifier and stage-specific few-shot prompting to produce accurate stage predictions. These stages are then connected into non-linear workflows using edge prediction, and stage properties are inferred from sub-utterance context. We compare CAG against strong single-prompt and agentic baselines, showing improved accuracy and efficiency—reducing token usage by over 60\%. Our architecture is modular, interpretable, and capable of end-to-end workflow generation, including robust validation steps. To our knowledge, this is the first system with a detailed evaluation across stage prediction, edge layout, and property generation for natural-language-driven ETL authoring.

Classifier-Augmented Generation for Structured Workflow Prediction

Ranking online reviews by their intrinsic quality is a critical task for e-commerce platforms and information services, impacting user experience and business outcomes. However, *quality* is a domain-dependent and dynamic concept, making its assessment a formidable challenge. Traditional methods relying on hand-crafted features are unscalable across domains and fail to adapt to evolving content patterns, while modern deep learning approaches often produce black-box models that lack interpretability and may prioritize semantics over quality. To address these challenges, we propose AutoQual, an LLM-based agent that automates the discovery of interpretable features for review quality. AutoQual mimics a human research process, iteratively generating feature hypotheses through reflection, operationalizing them via autonomous tool implementation, and accumulating experience in a persistent memory. This framework allows for the adaptive discovery of fine-grained and effective quality indicators. We deploy our method on a large-scale online platform with a billion-level user base. Large-scale A/B testing confirms its effectiveness, increasing average reviews viewed per user by 0.79\% and the conversion rate of review readers by 0.27\%.

AutoQual: An LLM Agent for Automated Discovery of Interpretable Features for Review Quality Assessment

Large language models (LLMs) can modify JSON documents through natural language commands, but current approaches regenerate entire structures for each edit, resulting in computational inefficiency. We present JSON Whisperer, a framework that enables LLMs to generate RFC 6902 diff patches-expressing only the necessary modifications-rather than complete documents.
We identify two key challenges in patch-based editing: (1) LLMs often miss related updates when generating isolated patches, and (2) array manipulations require tracking index shifts across operations, which LLMs handle poorly. To address these issues, we introduce EASE (Explicitly Addressed Sequence Encoding), which transforms arrays into dictionaries with stable keys, eliminating index arithmetic complexities.
Our evaluation shows that patch generation with EASE reduces token usage by 31\% while maintaining edit quality within 5\% of full regeneration with particular gains for complex instructions and list manipulations.

JSON Whisperer: Efficient JSON Editing with LLMs

Recently, the demand for small, efficient reasoning models to support real-world applications has driven the exploration of knowledge distillation approaches that balance reasoning performance and inference speed. In this paper, we further extend the DistilQwen model collection, initialized from Qwen models, by introducing four model series specifically designed to meet industrial needs. The distilled model collection includes: (1) slow-thinking models, optimized for reasoning tasks requiring high accuracy; (2) two series of adaptive-thinking models, which dynamically adjust their reasoning strategies based on input tasks to maximize efficiency across varied scenarios; and (3) distilled reward models for adaptive thinking, which support further reinforcement learning of reasoning models utilizing distilled knowledge. Comprehensive evaluations across several benchmarks demonstrate the inference efficiency and strong reasoning performance of reasoning models, together with the usefulness of distilled reward models. We further show how these models benefit industry practitioners by providing scalable model training and inference functionalities in an AI platform.

Downloads

Next from EMNLP 2025

Controllable Clustering with LLM-driven Embeddings

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EMNLP 2025

Controllable Clustering with LLM-driven Embeddings

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads