arXiv Computation and Language October 23, 2026 By Mullosharaf K. Arabov, Svetlana S. Khaybullina

Parameter-Efficient Adaptation of Large Language Models to a Low-Resource Agglutinative Language: A Corpus, Datasets, and a Reproducible Benchmark for the Bashkir Language

Read the original on arXiv Computation and Language →

The paper introduces the Bashkir Web Corpus, a 71,567‑document collection (46.9 M tokens) and three labeled datasets for text classification, and benchmarks eight large language models (from 82 M to 8 B parameters) on binary, multiclass, and multilabel tasks. It evaluates three adaptation strategies—full fine‑tuning, QLoRA, and in‑context learning—alongside continual pretraining, finding that QLoRA on 3.8–8 B models outperforms encoder‑only fine‑tuning for binary classification, while continual pretraining’s impact varies by model and task. The study also reports tokeniser fertility differences, shows that Cyrillic tokenisers pretrained on Russian perform best, and demonstrates that LLM‑generated ratings correlate with native judgments for grammar and coherence but not for language correctness, underscoring the need for native evaluation.

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 Computation and Language.

arXiv AI

Oct 1

Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation

arXiv:2605.07111v3 Announce Type: replace-cross Abstract: Recent literature on fine-tuning Large Language Models highlights a fundamental debate. While Full Fine-Tuning (FFT) provides greater represe...

By Haozhan Tang, Xiuqi Zhu, Xinyin Zhang, Boxun Li, Virginia Smith, Kevin Kuo

llms fine-tuning data-engineering

arXiv Computation and Language

Oct 8

sk-bench: A Native-First Benchmark for Evaluating Large Language Models in Slovak

arXiv:2610.09152v1 Announce Type: new Abstract: Multilingual LLM benchmarks omit Slovak, a morphologically rich West Slavic language of five million speakers, or cover it only by machine translation....

By Marek \v{S}uppa, Ivan Vykopal, Andrej Ridzik, Kristi\'an Sopkovi\v{c}, Nat\'alia K\v{n}a\v{z}ekov\'a, Jaroslav Kop\v{c}an, Miroslav Bl\v{s}t\'ak, Vikt\'oria Ondrejov\'a, Daniel Hl\'adek, Michal Gregor, Martin Tamajka, Mari\'an \v{S}imko

llms nlp benchmarks

arXiv AI

Sep 1

Capability-Stratified Degradation in Ternary Language Models

arXiv:2608.28809v1 Announce Type: new Abstract: Extreme low-bit inference offers a route toward smaller models and constrained deployment. Ternary language models restrict weights to $\{-1,0,+1\}$, a...

By Anirudh Malik, M Sparsh Mehra, Poojith Devan

llms fine-tuning benchmarks

arXiv Computation and Language

Oct 27

Task Format as a Factor in Adaptation: A Controlled Comparison of LoRA and Full Fine-Tuning for Prompt-Safety Classification

arXiv:2610.33813v1 Announce Type: new Abstract: Fine-tuning a language model for classification involves many design decisions. Comparisons of low-rank adaptation (LoRA) with full fine-tuning are com...

By John Alderete, Srirakshith Betageri, Jennifer Flynn, Alex Greenhalgh, Josh Rubin, Yoshita Varma Annam

llms fine-tuning safety

arXiv Machine Learning

Oct 28

Bug Severity Classification with Fine-Tuned Open LLMs: A Paired Comparison with TF-IDF and CodeBERT Baselines on Eclipse Reports

The study evaluates the effectiveness of small, fine‑tuned open large language models (LLMs) for predicting bug severity in Eclipse bug reports, comparing them against traditional TF‑IDF classifiers and a fine‑tuned CodeBERT encoder. Using a balanced sample of 20,000 reports, eleven model configurations—including zero‑shot and QLoRA‑fine‑tuned Llama 3 8B and Qwen 2.5 7B—were tested on 2,000 held‑out reports, revealing that fine‑tuned LLMs achieve macro‑F1 scores around 0.52, comparable to CodeBERT, while zero‑shot LLMs lag behind TF‑IDF baselines. The analysis also highlights class‑specific performance differences and the impact of duplicate reports on model evaluation.

By Asset Kanatov, Talgar Bayan, Askar Boranbayev

llms fine-tuning

arXiv Computation and Language

Oct 19

Pearl-Chat: Engineering a native Luganda language model from first principles in JAX and Flax NNX

Pearl-Chat is an 8.5‑million‑parameter causal decoder‑only transformer built from first principles in JAX and Flax NNX, specifically engineered for the low‑resource Bantu language Luganda. It uses a byte‑level BPE tokenizer trained from scratch on curated Luganda corpora, achieving a two‑fold compression improvement over larger models like LLaMA‑3 and Gemma. The model is trained in two stages—pre‑training on diverse Luganda text and instruction fine‑tuning on dialogues—reaching a validation perplexity of 1.20 and high ROUGE‑L and BLEU scores on a held‑out diagnostic benchmark, with all weights, tokenizers, and code released openly.

By Wesley Kambale, Roland Sankara

llms nlp fine-tuning benchmarks

One email a morning, machine-written

Related stories

Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation

sk-bench: A Native-First Benchmark for Evaluating Large Language Models in Slovak

Capability-Stratified Degradation in Ternary Language Models

Task Format as a Factor in Adaptation: A Controlled Comparison of LoRA and Full Fine-Tuning for Prompt-Safety Classification

Bug Severity Classification with Fine-Tuned Open LLMs: A Paired Comparison with TF-IDF and CodeBERT Baselines on Eclipse Reports

Pearl-Chat: Engineering a native Luganda language model from first principles in JAX and Flax NNX