China

In-Image Machine Translation (IIMT) aims to translate images containing texts from one language to another. Current research of end-to-end IIMT mainly conducts on synthetic data, with simple background, single font, fixed text position, and bilingual translation, which can not fully reflect real world, causing a significant gap between the research and practical conditions. To facilitate research of IIMT in real-world scenarios, we explore Practical In-Image Multilingual Machine Translation (IIMMT). In order to convince the lack of publicly available data, we annotate the PRIM dataset, which contains real-world captured one-line text images with complex background, various fonts, diverse text positions, and supports multilingual translation directions. We propose an end-to-end model VisTrans to handle the challenge of practical conditions in PRIM, which processes visual text and background information in the image separately, ensuring the capability of multilingual translation while improving the visual quality. Experimental results indicate the VisTrans achieves a better translation quality and visual effect compared to other models.

EMNLP 2025

PRIM: Towards Practical In-Image Multilingual Machine Translation

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.

Natural language transformation (NLT) tasks, such as machine translation (MT) and text style transfer (TST), require models to generate accurate and contextually appropriate outputs. However, existing approaches face significant challenges, including the computational costs of leveraging large pre-trained models and the limited generalization ability of fine-tuned smaller models. In this paper, we propose a novel framework that combines the flexibility of prompting with the cost-effectiveness of fine-tuning. Our method enhances smaller models by integrating In-Context Examples (ICE) retrieved from training data, enabling the model to better capture contextual information and align with user preferences. We further improve performance through hierarchical contrastive learning and dynamic preference inference mechanisms. Experimental results demonstrate that our approach outperforms existing methods, such as Supervised Fine Tuning (SFT), Direct Preference Optimization (DPO), and Contrastive Preference Optimization (CPO), across both MT and TST tasks, providing a more efficient solution for resource-constrained environments.

Tuning Less, Prompting More: In-Context Preference Learning Pipeline for Natural Language Transformation

Multilingual large language models (LLMs) often exhibit factual inconsistencies across languages, with significantly better performance in factual recall tasks in English than in other languages. The causes of these failures, however, remain poorly understood. Using mechanistic analysis techniques, we uncover the underlying pipeline that LLMs employ, which involves using the English-centric factual recall mechanism to process multilingual queries and then translating English answers back into the target language. We identify two primary sources of error: insufficient engagement of the reliable English-centric mechanism for factual recall, and incorrect translation from English back into the target language for the final answer. To address these vulnerabilities, we introduce two vector interventions, both independent of languages and datasets, to redirect the model toward better internal paths for higher factual consistency. Our interventions combined increase the recall accuracy by over 35 percent for the lowest-performing language. Our findings demonstrate how mechanistic insights can be used to unlock latent multilingual capabilities in LLMs.

Paths Not Taken: Understanding and Mending the Multilingual Factual Recall Pipeline

Recently, the rapid development of multilingual social media platforms (SNS) exacerbates new challenges in SNS content anomaly detection due to data islands and linguistic imbalance. While federated learning (FL) and parameter-efficient fine-tuning (PEFT) offer potential solutions in most cases, when every client is multilingual, existing solutions struggle with multilingual heterogeneity: 1) entangled language-specific knowledge during aggregation, 2) noise from minority languages, and 3) unstable cross-platform collaboration. Based on the asymmetric nature of LoRA, we propose MuLA-F, a multilingual Federated LoRA introducing SVD-based language-specific disentanglement of LoRA blocks and a local orthogonal tuning strategy. Evaluations across three SNS content anomaly detection tasks demonstrate MuLA-F’s superiority in multilingual performance while reducing multilingual knowledge conflicts and communication rounds.

Multilingual Federated Low-Rank Adaptation for Collaborative Content Anomaly Detection across Multilingual Social Media Participants

In this paper we present a comprehensive analysis of lexical semantic divergence between cognate words and borrowings in the Romance languages. We experiment with different algorithms for false friend detection including deceptive cognate and deceptive borrowings and correction and evaluate them systematically on cognate and borrowing pairs in the five Romance languages. We use the most complete and reliable dataset of cognate words based on etymological dictionaries for the five main Romance languages (Italian, Spanish, Portuguese, French and Romanian) to extract deceptive cognates and borrowings automatically based on usage, and freely publish the lexicon of obtained true and deceptive cognate and borrowings in every Romance language pair.

Friend or Foe? A Computational Investigation of Semantic False Friends across Romance Languages

We unveil that internal representations in large language models (LLMs) serve as reliable proxies of learned knowledge, and propose **RECALL**, a novel representation-aware model merging framework for continual learning without access to historical data. RECALL computes inter-model similarity from layer-wise hidden representations over clustered typical samples, and performs adaptive, hierarchical parameter fusion to align knowledge across models. This design enables the preservation of domain-general features in shallow layers while allowing task-specific adaptation in deeper layers. Unlike prior methods that require task labels or incur performance trade-offs, RECALL achieves seamless multi-domain integration and strong resistance to catastrophic forgetting. Extensive experiments across five NLP tasks and multiple continual learning scenarios show that RECALL outperforms baselines in both knowledge retention and generalization, providing a scalable and data-free solution for evolving LLMs.

RECALL: REpresentation-aligned Catastrophic-forgetting ALLeviation via Hierarchical Model Merging

We present Speech Vecalign, a novel parallel speech document alignment method that aligns speech segment embeddings and does not depend on text transcriptions. Compared to the baseline method Global Mining, a variant of speech mining, Speech Vecalign produces longer speech-to-speech alignments. It also demonstrates greater robustness than Local Mining, another speech mining variant, as it produces less noise. We applied Speech Vecalign to 3,000 hours of unlabeled parallel English-German (En-De) speech documents from VoxPopuli, yielding about 1,000 hours of high-quality alignments. We then trained En-De speech-to-speech translation models on the aligned data. Speech Vecalign improves the En-to-De and De-to-En performance over Global Mining by 0.37 and 0.18 BLEU, respectively. Moreover, our models match or outperform SpeechMatrix model performance, despite using 8 times fewer raw speech documents.

Speech Vecalign: an Embedding-based Method for Aligning Parallel Speech Documents

We introduce TurBLiMP, the first Turkish benchmark of linguistic minimal pairs, designed to evaluate the linguistic abilities of monolingual and multilingual language models (LMs). Covering 16 linguistic phenomena with 1000 minimal pairs each, TurBLiMP fills an important gap in linguistic evaluation resources for Turkish. In designing the benchmark, we give extra attention to two properties of Turkish that remain understudied in current syntactic evaluations of LMs, namely word order flexibility and subordination through morphological processes. Our experiments on a wide range of LMs and a newly collected set of human acceptability judgments reveal that even cutting-edge Large LMs still struggle with grammatical phenomena that are not challenging for humans, and may also exhibit different sensitivities to word order and morphological complexity compared to humans.

TurBLiMP: A Turkish Benchmark of Linguistic Minimal Pairs

Aligning large language models (LLMs) with human preferences is a central challenge for building reliable AI systems. Most existing alignment approaches rely on static signals, such as predefined principles or offline human annotations to guide model behavior toward a fixed approximation of human preferences. However, LLMs can exhibit distributional drift during training, and static alignment mechanisms lack the capacity to adaptively correct misaligned behaviors as they emerge. To address this limitation, we develop a two-stage framework that enables dynamic and continuous alignment. In the first stage, a constitution is continually revised based on observed model behaviors, and models are trained to comply with these evolving principles. In the second stage, this learned constitution is used to guide reinforcement learning, encouraging the model to align with the updated normative signals. We refer to this framework as COCOA: Co-evolution of Constitutions and AI Models. We show that COCOA enables a 7B model to greatly improve safety—raising StrongReject score from 0.741 to 0.935 and Safe-RLHF accuracy from 77.76% to 90.64% without human annotations, reaching performance close to much larger state-of-the-art models.

Governance in Motion: Co-evolution of Constitutions and AI models for Scalable Safety

Reward modeling is a key step in building safe foundation models when applying reinforcement learning from human feedback (RLHF) to align Large Language Models (LLMs). However, reward modeling based on the Bradley-Terry (BT) model assumes a global reward function, failing to capture the inherently diverse and heterogeneous human preferences. Hence, such oversimplification limits LLMs from supporting personalization and pluralistic alignment. Theoretically, we show that when human preferences follow a mixture distribution of diverse subgroups, a single BT model has an irreducible error. While existing solutions, such as fine-grained annotations via prompting or structured preference elicitation, help address this issue, they are costly and constrained by predefined attributes, failing to fully capture the richness of human values. In this work, we introduce MiCRo, a two-stage framework that enhances personalized preference learning by leveraging large-scale binary preference datasets without requiring explicit fine-grained annotations. In the first stage, MiCRo employs a mixture of preferences to model diverse human preferences, enabling a flexible representation of diverse value systems. In the second stage, MiCRo integrates an online routing strategy that dynamically adapts mixture weights based on specific context to resolve ambiguity, allowing for efficient and scalable preference adaptation with minimal additional supervision. Experiments on multiple preference datasets demonstrate that MiCRo effectively captures diverse human preferences and significantly improves personalized preference learning on downstream tasks.

MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning

Recent advancements in tool learning have enabled large language models (LLMs) to integrate external tools, enhancing their task performance by expanding their knowledge boundaries. However, relying on tools often introduces trade-offs between performance, speed, and cost, with LLMs sometimes exhibiting overreliance and overconfidence in tool usage. This paper addresses the challenge of aligning LLMs with their knowledge boundaries to make more intelligent decisions about tool invocation. We propose a multi-objective alignment framework that combines probabilistic knowledge boundary estimation with dynamic decision-making, allowing LLMs to better assess when to invoke tools based on their confidence. Our framework includes two methods for knowledge boundary estimation—consistency-based and absolute estimation—and two training strategies for integrating these estimates into the model’s decision-making process. Experimental results on various tool invocation scenarios demonstrate the effectiveness of our framework, showing significant improvements in tool efficiency by reducing unnecessary tool usage.

Downloads

Next from EMNLP 2025

Tuning Less, Prompting More: In-Context Preference Learning Pipeline for Natural Language Transformation

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES