arXiv:2609.07309v1 Announce Type: cross Abstract: Robustness evaluation of large language models (LLMs) remains a critical challenge, particularly in assessing their sensitivity to perturbations in i...
Manacá-1B is a 1.72‑billion‑parameter, open decoder‑only language model trained from scratch for Brazilian Portuguese, released with a fully containerized, reproducible training pipeline and complete logs. The authors evaluate it against nine open baselines on four Portuguese benchmarks, reporting standard errors and paired significance tests, and find that Manacá-1B outperforms smaller models on LAMBADA‑PT while remaining competitive on commonsense completion. They also uncover a tokenizer‑related evaluation pitfall that can drastically lower accuracy and provide a simple fix, releasing all code, logs, and corrected tokenizer for full reproducibility.
arXiv:2609. 17435v1 Announce Type: new Abstract: We submit M\'eTRON-FR, a 125M GPT-2 pretrained on 92.
EuroAlpaca presents a task‑preserving localisation pipeline that translates English instruction‑tuning data into 50 European languages while maintaining task‑critical constraints. The method uses field‑wise machine translation or reconstructs task‑equivalent target‑language instances, followed by validation of coherence and consistency. Experiments show that EuroAlpaca improves instruction‑following accuracy by 12.9% over a baseline and outperforms direct translation on ROUGE‑L and F‑BERT metrics.
arXiv:2610.01428v1 Announce Type: cross Abstract: Generalization in large language models (LLMs) is the ability to produce consistent and semantically stable outputs when the same input is expressed...
arXiv:2607. 20526v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in settings where fluent but incorrect answers can be costly.
Dialectal variation remains a major challenge for multilingual language models. Perturbation-based continued pre-training (CPT) has emerged as a promising approach to improving robustness, yet existing work largely evaluates individual perturbation strategies in isolation and provides limited insight into why they work.
The paper introduces a black‑box, inference‑time diagnostic for low‑resource Automatic Post‑Editing (APE) that distinguishes whether poor performance is due to insufficient training data or inconsistent training signals. By varying an edit‑distance penalty and analyzing the resulting TER‑vs‑λ curve and confidence‑based constraint ordering, the authors identify two failure modes—Binary Collapse and Confident Miscalibration—across multiple language pairs. The diagnostic also suggests practical next steps, such as applying a static constraint for immediate accuracy gains, and the authors release new English‑Sinhala and English‑Tamil APE datasets with accompanying code.
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:2610.27724v1 Announce Type: new Abstract: Dense linear probes assess task readability from language model (LM) representations, but do not directly reveal how many features suffice for predicti...
The paper investigates multilingual confidence calibration in large language models, revealing that non‑English languages are systematically less well calibrated than English. By analyzing internal representations, the authors find that late‑intermediate layers provide a more reliable confidence signal than the final layer, which is biased by English‑centric training. They propose training‑free methods such as Language‑Aware Confidence Ensemble (LACE) to adaptively select optimal layers per language, aiming to improve global equity and trustworthiness of LLMs.
Fine‑tuning large language models on parallel data can improve translation quality but also causes catastrophic forgetting of general capabilities. The study evaluates several forgetting‑mitigation methods—anchored to auxiliary data, model outputs, and base model parameters—using Llama 3.2 1B Instruct and Llama 3.1 8B Instruct on Arabic‑English and Spanish‑English translation tasks. Elastic Weight Consolidation best preserves general benchmark performance, yet only data mixing with control‑task examples maintains instruction‑following abilities such as formality and grammatical gender control, though these gains do not generalize to unseen prompts.