Austria

Scaling law builds the relationship between training computation and validation loss, enabling researchers to effectively predict the loss trending of models across different levels of computation. However, a gap still remains between validation loss and the model&#39;s downstream capabilities, making it untrivial to apply scaling law to direct performance prediction for downstream tasks. The loss typically represents a cumulative penalty for predicted tokens, which are implicitly considered to have equal importance. Nevertheless, our studies have shown evidence that when considering different training data distributions, we cannot directly model the relationship between downstream capability and computation or token loss. To bridge the gap between validation loss and downstream task capabilities, in this work, we introduce Capability Salience Vector, which decomposes the overall loss and assigns different importance weights to tokens to assess a specific meta-capability, aligning the validation loss with downstream task performance in terms of the model’s capabilities. Experiments on various popular benchmarks demonstrate that our proposed Capability Salience Vector could significantly improve the predictability of language model performance on downstream tasks.

ACL 2025

Capability Salience Vector: Fine-grained Alignment of Loss and Capabilities for Downstream Task Scaling Law

capability salience vector

downstream task prediction

scaling law

## 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." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

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

Scaling law builds the relationship between training computation and validation loss, enabling researchers to effectively predict the loss trending of models across different levels of computation. However, a gap still remains between validation loss and the model's downstream capabilities, making it untrivial to apply scaling law to direct performance prediction for downstream tasks. The loss typically represents a cumulative penalty for predicted tokens, which are implicitly considered to have equal importance. Nevertheless, our studies have shown evidence that when considering different training data distributions, we cannot directly model the relationship between downstream capability and computation or token loss. To bridge the gap between validation loss and downstream task capabilities, in this work, we introduce Capability Salience Vector, which decomposes the overall loss and assigns different importance weights to tokens to assess a specific meta-capability, aligning the validation loss with downstream task performance in terms of the model’s capabilities. Experiments on various popular benchmarks demonstrate that our proposed Capability Salience Vector could significantly improve the predictability of language model performance on downstream tasks.

technical paper

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Inductive biases are inherent in every machine learning system, shaping how models generalize from finite data. In the case of neural language models (LMs), debates persist as to whether these biases align with or diverge from human processing constraints. To address this issue, we propose a quantitative framework that allows for controlled investigations into the nature of these biases. Within our framework, we introduce 
$m$-local entropy—an information-theoretic measure derived from average lossy-context surprisal—that captures the local uncertainty of a language by quantifying how effectively the preceding symbols disambiguate the next symbol. In experiments on both perturbed natural language corpora and languages defined by probabilistic finite-state automata (PFSA), we show that languages with higher $m$-local entropy are more difficult for Transformer and LSTM LMs to learn. These results suggest that neural LMs, much like humans, are highly sensitive to the local statistical structure of a language.

Learnability on the Information-Theoretic Continuum: Inductive Bias for Information Locality in Neural Language Models

Large Language Models (LLMs) are increasingly utilized in autonomous decision-making, where they sample options from vast action spaces. However, the heuristics that guide this sampling process remain under-explored. We study this sampling behavior and show that this underlying heuristics resembles that of human decision-making: comprising a descriptive component (reflecting statistical norm) and a prescriptive component (implicit ideal encoded in the LLM) of a concept. We show that this deviation of a sample from the statistical norm towards a prescriptive component consistently appears in concepts across diverse real-world domains like public health, and economic trends. To further illustrate the theory, we demonstrate that concept prototypes in LLMs are affected by prescriptive norms, similar to the concept of normality in humans. Through case studies and comparison with human studies, we illustrate that in real-world applications, the shift of samples toward an ideal value in LLMs' outputs can result in significantly biased decision-making, raising ethical concerns.

A Theory of LLM Sampling: Part Descriptive and Part Prescriptive

To compare autoregressive language models at scale, we propose using log-likelihood vectors computed on a predefined text set as model features. This approach has a solid theoretical basis: when treated as model coordinates, their squared Euclidean distance approximates the Kullback-Leibler divergence of text-generation probabilities. Our method is highly scalable, with computational cost growing linearly in both the number of models and text samples, and is easy to implement as the required features are derived from cross-entropy loss. Applying this method to over 1,000 language models, we constructed a ``model map,'' providing a new perspective on large-scale model analysis.

Mapping 1,000+ Language Models via the Log-Likelihood Vector

Recent multilingual pretrained language models (mPLMs) often avoid using language embeddings -- learnable vectors assigned to individual languages. However, this places a significant burden on token representations to encode all language-specific information, which may hinder language neutrality. To address this limitation, we propose Language-Script Aware Multilingual Pretraining (LangSAMP), a method that incorporates both language and script embeddings to enhance representation learning. Specifically, we integrate these embeddings into the output of the Transformer blocks before passing the final representations to the language modeling head for prediction. We apply LangSAMP to the continual pretraining of XLM-R on a highly multilingual corpus covering more than 500 languages. The resulting model consistently outperforms the baseline in zero-shot crosslingual transfer across diverse downstream tasks. Extensive analysis reveals that language and script embeddings capture language- and script-specific nuances, which benefits more language-neutral representations, proven by improved pairwise cosine similarity. In our case study, we also show that language and script embeddings can be used to select better source languages for crosslingual transfer. We make our code and models publicly available at \url{https://github.com/cisnlp/LangSAMP}.

LangSAMP: Language-Script Aware Multilingual Pretraining

This paper addresses the critical need for democratizing large language models (LLM) in the Arab world, a region that has seen slower progress in developing models comparable to state-of-the-art offerings like GPT-4 or GPT-3.5, due to a predominant focus on mainstream languages (e.g., English and Chinese). One practical objective for Arabic LLMs is to utilize Arabic-specific vocabulary in the tokenizer to accelerate decoding. However, using a different vocabulary often leads to degradation of the model's learned knowledge, since many words become out-of-vocabulary (OOV) at the beginning of training. Inspired by the vocabulary learning during Second Language (Arabic) Acquisition for humans, the released AraLLaMA employs progressive vocabulary expansion, which is implemented by a modified BPE algorithm that progressively extends the Arabic subwords in its dynamic vocabulary during training, thereby balancing the OOV ratio at every stage. The ablation study demonstrated the effectiveness of Progressive Vocabulary Expansion.Moreover, AraLLaMA achieves decent performance comparable to the best Arabic LLMs across a variety of Arabic benchmarks. Our model weights are available at: {\url{https://github.com/FreedomIntelligence/AraLLaMa}}.

Second Language (Arabic) Acquisition of LLMs via Progressive Vocabulary Expansion

Large Language Models (LLMs) have shown remarkable capabilities in natural language processing but exhibit significant performance gaps among different languages. Most existing approaches to address these disparities rely on pretraining or fine-tuning, which are resource-intensive. To overcome these limitations without incurring significant costs, we propose Inference-Time Cross-Lingual Intervention (INCLINE), a novel framework that enhances LLM performance on low-performing (source) languages by aligning their internal representations with those of high-performing (target) languages during inference. INCLINE initially learns alignment matrices using parallel sentences from source and target languages through a Least-Squares optimization, and then applies these matrices during inference to transform the low-performing language representations toward the high-performing language space. Extensive experiments on nine benchmarks with five LLMs demonstrate that INCLINE significantly improves performance across diverse tasks and languages, compared to recent strong baselines. Our analysis demonstrates that INCLINE is highly cost-effective and applicable to a wide range of applications. In addition, we release the code to foster research along this line.

Bridging the Language Gaps in Large Language Models with Inference-Time Cross-Lingual Intervention

Language is a form of symbolic capital that affects people's lives in many ways (Bourdieu1977,1991). As a powerful means of communication, it reflects identities, cultures, traditions, and societies more broadly. Therefore, data in a given language should be regarded as more than just a collection of tokens. Rigorous data collection and labeling practices are essential for developing more human-centered and socially aware technologies. Although there has been growing interest in under-resourced languages within the NLP community, work in this area faces unique challenges, such as data scarcity and limited access to qualified annotators.

In this paper, we collect feedback from individuals directly involved in and impacted by NLP artefacts for medium- and low-resource languages. We conduct both quantitative and qualitative analyses of their responses and highlight key issues related to: (1) data quality, including linguistic and cultural appropriateness; and (2) the ethics of common annotation practices, such as the misuse of participatory research. Based on these findings, we make several recommendations for creating high-quality language artefacts that reflect the cultural milieu of their speakers, while also respecting the dignity and labor of data workers.

Building Better: Avoiding Pitfalls in Developing Language Resources when Data is Scarce

Automatic N-gram based metrics such as ROUGE are widely used for evaluating generative tasks such as summarization. While these metrics are considered indicative (even if imperfect), of human evaluation for English, their suitability for other languages remains unclear. To address this, in this paper we systematically assess evaluation metrics for generation — both n-gram-based and neural-based— to assess their effectiveness across languages and tasks. Specifically, we design a large-scale evaluation suite across eight languages from four typological families — agglutinative, isolating, low-fusional, and high-fusional — from both low- and high-resource languages, to analyze their correlations with human judgments. Our findings highlight the sensitivity of the evaluation metric to the language type at hand. For example, for fusional languages, n-gram-based metrics demonstrate a lower correlation with human assessments, compared to isolating and agglutinative languages. We also demonstrate that tokenization considerations can significantly mitigate this for fusional languages with rich morphology, up to reversing such negative correlations. Additionally, we show that neural-based metrics specifically trained for evaluation, such as COMET, consistently outperform other neural metrics and correlate better than ngrmas metrics with human judgments in low-resource languages. Overall, our analysis highlights the limitations of n-gram metrics for fusional languages and advocates for investment in neural-based metrics trained for evaluation tasks.

Beyond N-Grams: Rethinking Evaluation Metrics and Strategies for Multilingual Abstractive Summarization

Empirical risk minimization (ERM) incentivizes models to exploit shortcuts, i.e., spurious correlations between input attributes and labels that are prevalent in the majority of the training data but unrelated to the task at hand. This reliance hinders generalization on minority examples, where such correlations do not hold. Existing shortcut mitigation approaches are model-specific, difficult to tune, computationally expensive, and fail to improve learned representations. To address these issues, we propose InterpoLated Learning (InterpoLL) which interpolates the representations of majority examples to include features from intra-class minority examples with shortcut-mitigating patterns. This weakens shortcut influence, enabling models to acquire features predictive across both minority and majority examples. Experimental results on multiple natural language understanding tasks demonstrate that InterpoLL improves minority generalization over both ERM and state-of-the-art mitigation methods, without compromising accuracy on majority examples. Notably, these gains persist across encoder, encoder-decoder, and decoder-only architectures, demonstrating the method's broad applicability.

Mitigating Shortcut Learning with InterpoLated Learning

Reward models (RMs) play a crucial role in reinforcement learning from human feedback (RLHF), aligning model behavior with human preferences. However, existing benchmarks for reward models show a weak correlation with the performance of optimized policies, suggesting that they fail to accurately assess the true capabilities of RMs. To bridge this gap, we explore several evaluation designs through the lens of reward overoptimization, i.e., a phenomenon that captures both how well the reward model aligns with human preferences and the dynamics of the learning signal it provides to the policy. The results highlight three key findings on how to construct a reliable benchmark: (i) it is important to minimize differences between chosen and rejected responses beyond correctness, (ii) evaluating reward models requires multiple comparisons across a wide range of chosen and rejected responses, and (iii) given that reward models encounter responses with diverse representations, responses should be sourced from a variety of models. However, we also observe that a extremely high correlation with degree of overoptimization leads to comparatively lower correlation with certain downstream performance. Thus, when designing a benchmark, it is desirable to use the degree of overoptimization as a useful tool, rather than the end goal.

Downloads

Next from ACL 2025

Learnability on the Information-Theoretic Continuum: Inductive Bias for Information Locality in Neural Language Models

Similar lecture

Temporal Scaling Law for Large Language Models