arXiv AI October 26, 2026 By Giang Son Nguyen, Wray Buntine, Dung D. Le, Wenya Wang

"Not" is Not Lost in Translation: A Causal Analysis of Negation Representations in Multilingual LLMs

The study investigates how multilingual large language models (LLMs) handle negation across languages. Using causal interventions on synthetic and FLORES-200 derived datasets, the authors find that negation representations are causally controllable and largely shared across English, French, Vietnamese, and Chinese. They demonstrate that a single sparse autoencoder feature can control polarity across languages and that these representations generalize to additional languages.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

llms

arXiv Computation and Language

Sep 2

Separating Syntax from Language: A Mechanistic Account of Translation in Multilingual LLMs

arXiv:2609.01356v1 Announce Type: new Abstract: Multilingual large language models (mLLMs) achieve strong performance in machine translation, yet our understanding of the mechanisms by which they tra...

By Mikhail Sonkin, Tanja Baeumel, Daniil Gurgurov, Josef van Genabith, Simon Ostermann

llms nlp

arXiv AI

Sep 15

Not all Negation Cues are Equal: Affixal Negations Yield Better Negation Understanding

The paper introduces NegCue, a large-scale dataset of 1.8 million samples that includes single-word, multi-word, and affixal negation cues, totaling over 200 unique forms. The authors pre-train encoder-only language models and large language models on this dataset to study how different negation types influence understanding. Experiments on five downstream benchmarks reveal that affixal negations provide the most significant performance gains, whereas single-word negations yield modest improvements, and that additional pre-training benefits both model types.

By Tian Tan, Eduardo Blanco

llms benchmarks

arXiv Machine Learning

Aug 11

Embedding Initialization for Unseen Low-resource Languages in Multilingual NMT: A Case Study on Limbum-English Translation

arXiv:2608. 07629v1 Announce Type: cross Abstract: Multilingual neural machine translation models such as NLLB-200 cover 200 languages but leave thousands unsupported, including most Grassfields Bantu languages of Cameroon.

By Samiratu Ntohsi, Neza David Tuyishimire, Anesu Kafesu, Marvin Ogore, Samuel Oluwajunwonlo Babalola, Oche Ankeli

llms rag nlp fine-tuning

arXiv Machine Learning

Aug 4

LangFIR: Discovering Sparse Language-Specific Features from Monolingual Data for Language Steering

arXiv:2604. 03532v2 Announce Type: replace-cross Abstract: Large language models (LLMs) show strong multilingual capabilities, yet reliably controlling the language of their outputs remains difficult.

By Sing Hieng Wong, Hassan Sajjad, A. B. Siddique

llms benchmarks

Hugging Face Trending Papers

Aug 13

SAEVerbalizer: Generating Explanations for Sparse Autoencoder Features via Representation Verbalization

Sparse autoencoders (SAEs) are proposed to extract numerous features from large language model (LLM) representations, yet explaining these features still relies primarily on external observation. This reliance leads to superficial explanations inferred from observed model behavior and computational inefficiency from collecting such behavioral evidence at scale.

llms

arXiv Computation and Language

Sep 1

Generative vs. Encoder Models for Multilingual NER: A Comprehensive Empirical Study on Naamapadam

The paper compares generative and encoder-based neural models for multilingual Named Entity Recognition (NER) across the eleven languages of the Naamapadam benchmark. Five classic model families, four decoder-only large language models fine‑tuned with LoRA and 4‑bit NF4 quantisation, and nine generative models in zero‑to‑5‑shot inference were evaluated under strict CoNLL span‑level metrics. Encoder-based models (mBERT and XLM‑R) achieved substantially higher F1 scores—up to 0.675 on Hindi—than any generative architecture, with gaps of 7.5–40 percentage points; the best few‑shot result reached only 28% of the encoder baseline. The study identifies three language clusters (encoder‑dominant, partial‑coverage, and failure‑zone) and offers deployment guidelines based on transfer learning and low‑resource NLP principles.

By Jakkala Mahesh, Jatavath Shravan Kumar, Komalla Shivani, Sujoy Sarkar

llms nlp fine-tuning benchmarks