🏆We are proud to announce that the Computational Linguistics Journal impact factor is now 13.4! This ranks it 13 out of 210 journals in Computer Science-Artificial Intelligence, 7 out of 185 in Computer Science-Interdisciplinary Applications, 1️⃣out of 312 in Linguistics! #NLProc #NLP

CL Journal has over 50 years of history in the field. While the field is moving, some of the articles from decades ago provide perspective and still relevant content. What do you think of these articles from 20 and 40 years ago? (2006 and 1986) & yes, they used to come in paper format too! #NLProc

While we're waiting for the next issue of the CL Journal, you can get early access to upcoming articles here: direct.mit.edu/coli/online-... #NLProc #CLJournal

Like humans, LLMs can be right for the wrong reasons. They can also be wrong for the wrong reasons. Anthropocentric bias has gone largely unexamined, posing a serious obstacle to the objective assessment of LLM capacities. Read more at doi.org/10.1162/COLI... #NLProc #CLJournal @raphaelmilliere.com

Interpretability provides a toolset for understanding how and why LMs behave in certain ways. This survey proposes a perspective on interpretability research grounded in causal mediation analysis: doi.org/10.1162/COLI... #NLProc #CLJournal @jannikbrinkmann.bsky.social @amuuueller.bsky.social

How Can We Effectively Expand the Vocabulary of LLMs with 0.01GB of Target Language Text? This article explores a very important question for low-resource languages by experimenting with various techniques across 10 languages. A must-read at:https://doi.org/10.1162/COLI.a.581 @gucciiiii.bsky.social

Should NLP metrics for bilingual code-switching use words as the token level or Intonation Units? Authors of this article show that intonation units will enhance comparisons between bilingual individuals, settings, and communities: doi.org/10.1162/COLI... #NLProc #NLP @pennlinguistics.bsky.social

Have you heard of soft metrics such as soft micro F1? In this article, the authors argue that for evaluating model predictions with human label variation, the standard metrics may not be sufficient. Read more at: doi.org/10.1162/COLI... @kmkurn.sigmoid.social.ap.brid.gy #NLProc #NLP

To assess whether multilingual NLP performs well across languages, we need to evaluate it on all world languages.That is not feasible.This paper implements two sampling methods from linguistic typology and provides a Python package to facilitate this: doi.org/10.1162/COLI... @andreashhp.bsky.social

Developing methods to assess the factuality of LLMs has become urgent. This paper presents LLM-OASIS for the factuality evaluation task. It turns out it significantly challenges SOTA LLMs. If you're up for improving over what's out there, start reading: doi.org/10.1162/COLI... #CLjournal #NLProc

Language Models are susceptible to adversarial attacks, where even subtle perturbations to input texts adversely affect model performance. Yang et al. propose a novel method to tackle this called Defensive Dual Masking (DDM). Read more at: doi.org/10.1162/COLI... #NLProc #NLP

Do current LLMs with fast-improving functional linguistic abilities exhibit distinct localization of formal (e.g., producing fluent, grammatical text) and functional (e.g., reasoning and consistent fact retrieval) linguistic mechanisms? Answer in: doi.org/10.1162/COLI... @michaelwhanna.bsky.social

Have you heard of linguistic steganography? It seeks to conceal secret information within natural language text. Liu et al. propose a novel method called SA-ANS. It's a self-adaptive framework based on a self-adjusting Asymmetric Numeral System: doi.org/10.1162/COLI... #NLProc #NLP

If your research is around metaphor detection and interpretation, this article is for you! It introduces Meta4XNLI, the first parallel dataset for Natural Language Inference (NLI), in both English and Spanish: doi.org/10.1162/COLI... #NLProc @ragerri.bsky.social

Volume 52, Issue 1 of Computational Linguistics is released 📣 You can access this issue at direct.mit.edu/coli/issue/5... The articles included in this issue will be introduced here and on X x.com/CompLingJour...! #NLProc #NLP

What % of the NLP papers measure their impact in the real world? This paper proposes an impact evaluation of NLP models or systems for real-world usage, changing the research culture of NLP to focus more on real-world impact and less on SOTA-chasing: doi.org/10.1162/COLI... @ehudreiter.bsky.social

Hallucinations pose a substantial challenge to the reliability of LLMs in real-world scenarios. Zhang et al. survey methods of detection, explanation, & mitigation of hallucination, & provide a taxonomy & list of benchmarks for evaluation in this paper: doi.org/10.1162/COLI... @fredashi.bsky.social

Generative AI has advanced. But, for complex problems, it's lagging. In this position paper, the authors propose Human–AI Co-Construction (HAI-Co2), a framework for human–AI cooperative problem solving that facilitates such interaction. Read more at: doi.org/10.1162/COLI... @igurevych.bsky.social

Tasks such as summarization, QA & timeline creation need temporal expression normalization. The authors propose a novel method that deals with the known problems of temporal expression normalization on data scarcity, language and domain adaptation: doi.org/10.1162/COLI... @jmartinezromo.bsky.social

Have you heard of BLiMP for evaluating English LMs? Well, the authors of this paper introduce the Dutch version: BLiMP-NL. It's a benchmark set of linguistic minimal pairs for grammatical evaluation of Dutch LMs. Read more at: doi.org/10.1162/COLI... @stefanfrank.bsky.social @illc-uva.bsky.social

Interested in the semantic neighborhood of words? LGDE is a graph-based, local-diffusion framework for the discovery of keywords that are semantically similar to a seed dictionary. The authors release their code and data for a hate speech dataset too: doi.org/10.1162/COLI... #NLP

Tokenisation and Finite-State Transducers are two well-known concepts in computing. This article combines them for neural networks. Cognetta & Okazaki introduce a finite-state transduction framework that can encode all possible tokenizations of a regular language: doi.org/10.1162/COLI... #NLProc

Are you looking for measuring "syntactic diversity"? Martin, in this article, proposes derivational entropy rate, a metric that links grammar’s derivational entropy and the mean length of the utterances (MLU). Check it out: doi.org/10.1162/COLI... #NLProc

Private attributes of writers, such as age or native language, can be detectable. The authors of this article propose IDT (Interpretable Dual-Task), a method based on adversarial attacks for rewriting text to prevent detection of such private attributes: doi.org/10.1162/COLI... #NLP #NLProc

Tokenisation and Finite-State Transducers are two well-known concepts in computing. This article comes them for neural networks. They introduce a finite-state transduction framework that can encode all possible tokenizations of a regular language: doi.org/10.1162/COLI... #NLProc #NLP

Volume 51, Issue 4 of Computational Linguistics was released in Dec 2025 📣 If you haven't seen it yet, you can access it at direct.mit.edu/coli/issue/5... #NLProc #NLP

LLMs transformed our field. This article addresses (1) ways to better understand how LLMs work & (2) ways to rethink how the next generation of models should use data at scale. Authors also propose SILO to advance responsible data use: doi.org/10.1162/COLI... @sewonm.bsky.social #NLP #NLProc

NLP relies on Linguistics & that's capital RELIES! It encapsulates 6 major facets where linguistics contributes to NLP: Resources, Evaluation, Low-resource settings, Interpretability, Explanation, & the Study of language. Read it at doi.org/10.1162/coli... @nlopitz.bsky.social @complingy.bsky.social

Have you read the book "Automated Essay Scoring" by Beata Beigman Klebanov and Nitin Madnani? The review of this book by Anaïs Tack could give you a glimpse of this great book: doi.org/10.1162/coli... #NLProc #NLP @haikumana.bsky.social

What is cultural awareness in LLMs & what has been done in the literature for creating cross-cultural datasets, strategies for cultural inclusion & methodologies that have been used for benchmarking cultural awareness? This survey answers all these questions: doi.org/10.1162/COLI... #NLProc

Is there a relationship between bias and fundamental properties of LLMs? While LLMs are powerful, they contain harmful biases that can emerge unpredictably in their behavior. Read more on this in this thought-provoking position paper: doi.org/10.1162/coli... #NLProc #NLP

Hierarchical topic models (HTM) depend heavily on data representation and topic evaluation. Viegas et al. demonstrate that HTMs, by nature, demand novel ways of evaluating the quality of topics. Interested? Read it at: doi.org/10.1162/coli... #NLPRoc

An unsupervised approach to chunking removes the need for manual annotations. This drives the paper proposing a Hierarchical Recurrent Neural Network (HRNN) designed to model word-to-chunk and chunk-to-sentence compositions. Want to know more? Read it here: doi.org/10.1162/coli... #NLProc

Do subword tokens influence how LLMs represent the meanings of words and characters? Do LLMs “know” about the composition of Chinese characters? Do they recognize and utilize correspondences between radicals and initial tokens? Answers to these questions at doi.org/10.1162/coli... #NLProc

Argument structure analysis helps to understand human argumentation. Bao et al. present a unified generative framework for this: UniASA. It can uniformly address multiple argument structure analysis tasks in a seq2seq manner. Read about it here: doi.org/10.1162/coli... #NLProc @aixinsg.bsky.social

The primary goal of an adversarial attack is to change a model’s prediction. Can we automatically detect adversarial sentences? Tonni et al introduce a novel dataset of Likert-scale human evaluations on the suspiciousness of adversarial sentences. More at: doi.org/10.1162/coli... #NLProc #NLP

To what extent are widely used word representation models able to capture idiomaticity? Is this ability affected by the degree of idiomaticity of the noun compounds, the informativeness of the contexts or the languages involved? Answers in doi.org/10.1162/coli... #NLProc @alinevillav.bsky.social

Can Code-LLMs acquire better causal reasoning abilities? Do code prompts better describe the causal structure than text prompts? Liu et al. explore how effective complex reasoning can be accomplished with code prompts. Read more at: doi.org/10.1162/coli... #CodeLLMs #NLProc #NLP

Arabic text representation doesn't need to follow classic Arabic NLP! Ancient Arabic was dotless. How to imagine? Well, the authors of this article propose to use dotless representation successfully. Read about it at: doi.org/10.1162/coli... #NLProc #ArabicNLP #NLP

Do texts generated by language models refer? Ostertag, in "Language Models and Externalism: A Reply to Mandelkern and Linzen” challenges the previous answer to this question using ideas from philosophy (contrastive explanation). Read more at: doi.org/10.1162/coli... #NLProc

What does the next chapter in the NLP research look like? Despite substantial advances in NLP, esp. LLMs, they still struggle with "social awareness". Is social awareness what the NLP field should move into in the coming years? Read more at: aclanthology.org/2025.cl-2.10... #NLProc #NLP

Do LLMs have personalities? This paper introduces LMLPA, a framework for assessing linguistic personalities of LLMs using adapted Big Five inventories + AI raters, validated for reliability and consistency. 📄 Read in Computational Linguistics: direct.mit.edu/coli/article...

Evaluating NLG remains challenging — even with LLMs. This survey reviews LLM-based evaluation methods: prompting, fine-tuning, and human–LLM collaboration, and outlines key open problems. 📄 Read in Computational Linguistics: direct.mit.edu/coli/article...

LLMs can generate fluent text — but can they twist your tongue? TwisterLister generates phoneme-aware tongue twisters, with a 17k-example dataset and phonologically constrained decoding. 📄 Read the paper in Computational Linguistics: direct.mit.edu/coli/article...

Can LLMs learn possible human languages but struggle with impossible ones? Kallini et al. claimed evidence, but Hunter argues their key experiment conflates factors — the core question remains open. 📄 Read the article in Computational Linguistics: direct.mit.edu/coli/article...

Even state-of-the-art MT systems struggle with ambiguous words — especially rare senses. DiBiMT is a new fully human-curated benchmark across 8 language pairs to study disambiguation bias in MT. 📄 Read the paper in Computational Linguistics: direct.mit.edu/coli/article...

We're honored to announce the recipient of the ACL 2025 Distinguished Service Award! 🏅 This award recognizes extraordinary and sustained contributions to the computational linguistics community.

📚The dissertation will be published in the CL journal. 💬Catch the talk right after the award presentation! #NLProc Congratulations to all these outstanding dissertations in NLP and CL — the future of #NLProc is in excellent hands! 🙌

Sewon Min 🏆 presenting the essence of her thesis: Rethinking Data Use in Large Language Models -- It won the first ACL Computational Linguistics Doctoral Dissertation Award 🎓✨ #ACL2025NLP #CLJournal @complingjournal.bsky.social

📚Tom Sherborne: Modeling Cross-lingual Transfer for Semantic Parsing Sherborne’s dissertation develops sophisticated methods for cross-lingual transfer into low-resource languages, demonstrating their effectiveness in the context of semantic parsing for integration with database APIs.