China

Results reported in large-scale multilingual evaluations are often fragmented and confounded by factors such as target languages, differences in experimental setups, and model choices. We propose a framework that disentangles these confounding variables and introduces three interpretable metrics—the performance realisation ratio, its coefficient of variation, and language potential—enabling a finer-grained and more insightful quantification of actual performance disparities across both (i) models and (ii) languages. Through a case study of 13 model variants on 11 multilingual datasets, we demonstrate that our framework provides a more reliable measurement of model performance and language disparities, particularly for low-resource languages, which have so far proven challenging to evaluate. Importantly, our results reveal that higher overall model performance does not necessarily imply greater fairness across languages.

EMNLP 2025

Quantifying Language Disparities in Multilingual Large Language Models

meta-analysis of benchmarks

language disparities

multilingual evaluation

evaluation framework

EMNLP 2021 is planned to be a hybrid event in Punta Cana, Dominican Republic, with both on-site and fully virtual participation possible. The experience for on-site participants would closely approximate a normal pre-COVID *ACL conference, with 5-6 thematically organized parallel sessions and live Q/A and interactive discussion immediately after the talks. Presentations by virtual participants will be equitably interleaved with those of on-site participants, projected on the auditorium screens as if on-site, and also followed immediately by live Q/A and interactive discussion at a time during reasonable waking hours for the virtual presenter. For all participants, on-site and virtual, who are unable to attend a session due to either time-zone issues or because they are participating in another session live, talk recordings and slides will be available online at a minimum after the live presentation (and in many cases before as well), and questions may be submitted in advance on session-specific discussion boards and answered live in session with the usual visual aids if desired.

<iframe style="width:700px;height:400px" src="https://online.fliphtml5.com/ebtyf/ceby/"  seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe>

Please Note: The EMNLP registration system is not currently connected to the underline site as we are still in the process of building out EMNLP 2021. You will receive access instructions from underline the week of November 1st. 

Access is given only to EMNLP upon registration, if you have not registered please do so [here](https://2021.emnlp.org/registration).

Registered attendees will receive access the week of November 1st.

EMNLP 2021

EMNLP 2021 is planned to be a hybrid event in Punta Cana, Dominican Republic, with both on-site and fully virtual participation possible.

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*





<br>

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

A key challenge in MT evaluation is the inherent noise and inconsistency of human ratings. Regression-based neural metrics struggle with this noise, while prompting LLMs shows promise at system-level evaluation but performs poorly at segment level. In this work, we propose ReMedy, a novel MT metric framework that reformulates translation evaluation as a reward modeling task. Instead of regressing on imperfect human ratings directly, ReMedy learns relative translation quality using pairwise preference data, resulting in a more reliable evaluation. In extensive experiments across WMT22-24 shared tasks (39 language pairs, 111 MT systems), ReMedy achieves state-of-the-art performance at both segment- and system-level evaluation. Specifically, ReMedy-9B surpasses larger WMT winners and massive closed LLMs such as MetricX-13B, XCOMET-Ensemble, GEMBA-GPT-4, PaLM-540B, and finetuned PaLM2. Further analyses demonstrate that ReMedy delivers superior capability in detecting translation errors and evaluating low-quality translations.

ReMedy: Learning Machine Translation Evaluation from Human Preferences with Reward Modeling

Large language models (LLMs) have shown strong potential in enhancing text clustering when combined with traditional embedding models. However, existing methods predominantly treat LLMs as static pseudo-oracles, i.e., unidirectionally querying them for similarity assessment or data augmentation, while never seeking feedback from embedding models to improve them. In this work, we propose a training framework that enables bidirectional refinement between LLMs and embedding models. We first design task-aware prompts to guide the LLM in generating interpretations for input texts. These interpretations are projected into the embedding space, in which interpretations that are preferred by the embedding model are selected based on their distribution densities. The selected interpretations are then used to fine-tune the LLM via preference optimization to prioritize the generation of helpful interpretations. Meanwhile, we enhance the embedding model via contrastive learning on the generated interpretations and perform clustering on the output embeddings, leading to iterative co-training between the LLM and embedding model. Experiments on 14 benchmark datasets across 5 tasks demonstrate the effectiveness of our method.

Co-Evolving LLMs and Embedding Models via Density-Guided Preference Optimization for Text Clustering

Recent advancements in large language models (LLMs) showcase varied multilingual capabilities across tasks like translation, code generation, and reasoning. Previous assessments often limited their scope to fundamental natural language processing (NLP) or isolated capability-specific tasks. To alleviate this drawback, we aim to present a comprehensive multilingual multitask benchmark. First, we introduce P-MMEval, a large-scale benchmark covering fundamental and capability-specialized datasets. Furthermore, P-MMEval delivers consistent language coverage across various datasets and provides parallel samples. Finally, we conduct extensive experiments on representative multilingual model series to compare performances across models and tasks, explore the relationship between multilingual performances and factors such as tasks, model sizes, languages, and prompts, and examine the effectiveness of knowledge transfer from English to other languages. The resulting insights are intended to offer valuable guidance for future research.

P-MMEval: A Parallel Multilingual Multitask Benchmark for Consistent Evaluation of LLMs

Text generated by Large Language Models (LLMs) now rivals human writing, raising concerns about its misuse. However, mainstream AI-generated text detection (AGTD) methods primarily target document-level long texts and struggle to generalize effectively to sentence-level short texts. And current sentence-level AGTD (S-AGTD) research faces two significant limitations: (1) lack of a comprehensive evaluation on complex human-AI hybrid content, where human-written text (HWT) and AI-generated text (AGT) alternate irregularly, and (2) failure to incorporate contextual information, which serves as a crucial supplementary feature for identifying the origin of the detected sentence. Therefore, in our work, we propose \textbf{AutoFill-Refine}, a high-quality synthesis strategy for human-AI hybrid texts, and then construct a dedicated S-AGTD benchmark dataset. Besides, we introduce \textbf{SenDetEX}, a novel framework for \underline{s}entence-level AI-g\underline{en}erated text \underline{det}ection via styl\underline{e} and conte\underline{x}t fusion. Extensive experiments demonstrate that SenDetEX significantly outperforms in detection accuracy while exhibiting remarkable transferability and robustness among all baseline models.

SenDetEX: Sentence-Level AI-Generated Text Detection for Human-AI Hybrid Content via Style and Context Fusion

Out-of-distribution (OOD) detection remains challenging in text-rich networks, where textual features intertwine with topological structures. Existing methods primarily address label shifts or rudimentary domain-based splits, overlooking the intricate textual-structural diversity. For example, in social networks, where users represent nodes with textual features (name, bio) while edges indicate friendship status, OOD may stem from the distinct language patterns between bot and normal users. To address this gap, we introduce the TextTopoOOD framework for evaluating detection across diverse OOD scenarios: (1) attribute-level shifts via text augmentations and embedding perturbations; (2) structural shifts through edge rewiring and semantic connections; (3) thematically-guided label shifts; and (4) domain-based divisions. Furthermore, we propose TNT-OOD to model the complex interplay between Text aNd Topology using: 1) a novel cross-attention module to fuse local structure into node-level text representations, and 2) a HyperNetwork to generate node-specific transformation parameters. This aligns topological and semantic features of ID nodes, enhancing ID/OOD distinction across structural and textual shifts. Experiments on 11 datasets across four OOD scenarios demonstrate the nuanced challenge of TextTopoOOD for evaluating OOD detection in text-rich networks. Code will be released upon acceptance.

Text Meets Topology: Rethinking Out-of-distribution Detection in Text-Rich Networks

Recent major milestones have successfully reconstructed natural language from non-invasive brain signals (e.g. functional Magnetic Resonance Imaging (fMRI) and Electroencephalogram (EEG)) across subjects. However, we find current dataset splitting strategies for cross-subject brain-to-text decoding are wrong. Specifically, we first demonstrate that all current splitting methods suffer from data leakage problem, which refers to the leakage of validation and test data into training set, resulting in significant overfitting and overestimation of decoding models. In this study, we develop a right cross-subject data splitting criterion without data leakage for decoding fMRI and EEG signal to text. Some SOTA brain-to-text decoding models are re-evaluated correctly with the proposed criterion for further research.

Rethinking Cross-Subject Data Splitting for Brain-to-Text Decoding

Multilingual large language models (LLMs) possess impressive multilingual understanding and generation capabilities. However, their performance and cross-lingual alignment often lag for non-dominant languages. A common solution is to fine-tune LLMs on large-scale and more balanced multilingual corpus, but such approaches often lead to imprecise alignment and suboptimal knowledge transfer, struggling with limited improvements across languages. In this paper, we propose AlignX to bridge the multilingual performance gap, which is a two-stage representation-level framework for enhancing multilingual performance of pre-trained LLMs. In the first stage, we align multilingual representations with multilingual semantic alignment and language feature integration. In the second stage, we stimulate the multilingual capability of LLMs via multilingual instruction fine-tuning. Experimental results on several pre-trained LLMs demonstrate that our approach enhances LLMs' multilingual general and cross-lingual generation capability. Further analysis indicates that AlignX brings the multilingual representations closer and improves the cross-lingual alignment.

AlignX: Advancing Multilingual Large Language Models with Multilingual Representation Alignment

Logit-based LLM watermarking traces and verifies AI-generated content by maintaining green and red token lists and increasing the likelihood of green tokens during generation. However, it struggles in low-entropy scenarios, where predictable outputs make green token selection difficult without disrupting natural text flow. Existing approaches address this by assuming access to the original LLM to calculate entropy and selectively watermark high-entropy tokens. However, these methods face two major challenges: (1) high computational costs and detection delays due to reliance on the original LLM, and (2) potential risks of model leakage. To address these limitations, we propose Invisible Entropy (IE), a watermarking paradigm designed to enhance both safety and efficiency. Instead of relying on the original LLM, IE introduces a lightweight feature extractor and an entropy tagger to predict whether the entropy of the next token is high or low. Furthermore, based on theoretical analysis, we developed a threshold navigator that adaptively sets entropy thresholds. It identifies a threshold where the watermark ratio decreases as the green token count increases, enhancing the naturalness of the watermarked text and improving detection robustness. Experiments on HumanEval and MBPP datasets demonstrate that IE reduces parameter size by 99% while achieving performance on par with state-of-the-art methods: https://anonymous.4open.science/r/IE-Official.

Invisible Entropy: Towards Safe and Efficient Low-Entropy LLM Watermarking

Accurately evaluating machine-translated text remains a long-standing challenge, particularly for long documents. Recent work has shown that large language models (LLMs) can serve as reliable and interpretable sentence-level translation evaluators via MQM error span annotations. With modern LLMs supporting larger context windows, a natural question arises: can we feed entire document translations into an LLM for quality assessment? Ideally, evaluation should be invariant to text length, producing consistent error spans regardless of input granularity. However, our analysis shows that text length significantly impacts evaluation: longer texts lead to fewer error spans and reduced system ranking accuracy. To address this limitation, we evaluate several strategies, including granularity-aligned prompting, Focus Sentence Prompting (FSP), and a fine-tuning approach to better align LLMs with the evaluation task. The latter two methods largely mitigate this length bias, making LLMs more reliable for long-form translation evaluation.

Same evaluation, more tokens: On the effect of input length for machine translation evaluation using Large Language Models

Transformer-based Large Language Models (LLMs) struggle with inputs exceeding their training context window due to positional out-of-distribution (O.O.D.) issues that disrupt attention. Existing solutions, including fine-tuning and training-free methods, face challenges like inefficiency, redundant interpolation, logit outliers, or loss of local positional information. We propose Greedy Attention Logit Interpolation (GALI), a training-free method that improves length extrapolation by greedily reusing pretrained positional intervals and interpolating attention logits to eliminate outliers. GALI achieves stable and superior performance across a wide range of long-context tasks without requiring input-length-specific tuning. Our analysis further reveals that LLMs interpret positional intervals unevenly and that restricting interpolation to narrower ranges improves performance, even on short-context tasks. GALI represents a step toward more robust and generalizable long-text processing in LLMs.

Downloads

Next from EMNLP 2025

ReMedy: Learning Machine Translation Evaluation from Human Preferences with Reward Modeling

Similar lecture

Language Clustering for Multilingual Named Entity Recognition

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES