The Flow has not summarised this story yet — read it at arXiv Computation and Language.
The paper introduces a layered taxonomy for annotating grammatical errors in Chinese learner writing, aiming for consistency and linguistic relevance. It first classifies character- and punctuation-level orthographic errors by edit operation and subtype, then assigns other errors a three-layer core label that combines edit operation, linguistic domain, and part of speech, with optional Chinese-specific extensions. The taxonomy is evaluated through coverage analysis of automatically extracted edits and a preliminary consistency study using five large language models, confirming the layered approach while highlighting areas needing refinement.
The paper presents an end‑to‑end sequence‑to‑sequence approach for correcting Tamil spelling and grammar errors, leveraging progressively fine‑tuned transformer models (mT5‑small and mBART‑50). Using a synthetic corpus of 657,720 noisy‑clean sentence pairs across ten error categories, the authors introduce a four‑stage training schedule that targets surface noise, contextual grammar, single‑site sandhi, and multi‑site cross‑word sandhi. The best model, mBART‑50 v5, achieves 69.3% exact‑match accuracy on a balanced diagnostic set, with notable gains in sandhi (87.5%) and subject‑verb agreement (43.5%) accuracy, while also revealing a precision‑recall trade‑off for sandhi corrections.
Bangla-English tutoring requires more than producing a correct translation: learners also need explanations of grammar differences, awareness of their likely errors, and targeted practice. We present...
arXiv:2608. 03803v1 Announce Type: cross Abstract: Multilingual language models are deployed across a hundred or more languages, yet most benchmarks test whether a model can perform a task _in_ a language rather than whether it commands the language itself, conflating fluency with proficiency.
arXiv:2608.28776v1 Announce Type: new Abstract: Multilingual sentence embeddings are increasingly used to estimate semantic similarity across languages, yet their sensitivity to fine-grained translat...
SWORD is a new benchmark that tests large language models’ ability to reject factually incorrect statements across eight major languages by distorting Wikidata triples. The benchmark reveals that models often perform better on semantically plausible distortions than on random ones, indicating a reliance on distributional familiarity rather than true factual verification. It also shows significant performance drops for East Asian languages, with gaps up to 28 percentage points, highlighting asymmetric multilingual factual reasoning capabilities.