Jul 2026· Annual International ACM SIGIR Conference on Research and Development in Information Retrieval· 0 citations· 41 references
Computer Science
TL;DR
Evidence of cross-lingual efficacy of code-based LLMs for Chinese QA tasks, further enhanced through Code Llama-M's expanded Chinese vocabulary is found, and successful application of the fine-tuned LLM in a live assistant system, enhancing user experience is demonstrated.
Abstract
This paper explores advancements in automated Question-Answer (QA) extraction using large language models (LLMs), addressing challenges in transforming unstructured text into high-quality, retrievable QA pairs. Traditional approaches, whether through segmented question and answer generation or end-to-end extraction, often struggle with efficiency, dataset limitations, and performance consistency. Leveraging recent progress in LLMs, we constructed a large-scale Chinese QA extraction dataset with 143,846 documents and evaluated multiple fine-tuned models on public and private datasets. Surprisingly, code-based English LLMs outperformed Chinese-specialized models on Chinese text with a lower hallucination rate. Building upon this finding, we enhanced the best-performing code-based model with an expanded Chinese vocabulary, creating Code Llama-M, which achieved better results. Integrating Code Llama-M into our internal assistant, Luo Ying, demonstrated notable user satisfaction gains, affirming its practical impact. Key contributions include: (i) creation of a robust Chinese QA extraction instruction dataset; (ii) evidence of cross-lingual efficacy of code-based LLMs for Chinese QA tasks, further enhanced through Code Llama-M's expanded Chinese vocabulary; and (iii) successful application of the fine-tuned LLM in a live assistant system, enhancing user experience.
BEAR-Bench (Bilingual Enterprise and Academic Reasoning), a self-contained, complex English-and-Russian benchmark comprising 1000 human-annotated questions based on text-rich business and scientific documents, is introduced, and existing hallucination detection methods are compared.
L. Chubarova, A. Kuleshova, D. P. Volkov et al.· 0 citations
Whether preserving the full recursive structure of user forum threads during post-training is a more effective first step toward knowledge extraction than flattened question-answer pairs is investigated andEncoder–decoder architectures with bidirectional cross-attention are identified as a promising next step for exploiting the full collaborative structure of forum discourse.
Jeffrey D. Vitale· Machine Learning and Knowled...· 0 citations
Chinese Spelling Correction (CSC) is a fundamental task in Natural Language Processing (NLP) aimed at identifying and correcting character errors in Chinese texts. It significantly enhances text readability and semantic accuracy. Most deep learning-based CSC methods focus on isometric correction, ensuring identical lengths for input and output sequences. However, they struggle with variable-length errors like splitting errors—where a single character is incorrectly divided into two (e.g., splitting “明” into “日” and “月”). These errors are challenging because they disrupt token alignment, preventing standard sequence-labeling models from mapping inputs to outputs effectively. To overcome this limitation, we propose KSEC (Knowledge-enhanced Splitting Error Corrector), a novel framework tailored for variable-length corrections. KSEC automatically constructs a splitting character knowledge base from public corpora to provide factual validation for correction outcomes. Furthermore, we design a variable-length architecture integrating an attention mechanism and introduce an alignment-aware loss function that optimizes sequence-to-sequence token mapping. Extensive experiments on standard CSC and CSEC benchmarks demonstrate that KSEC achieves state-of-the-art performance among lightweight models of similar size and outperforms existing methods across multiple evaluation metrics.
Jiahao Wang, Guimin Huang, Yabing Wang et al.· Electronics· 0 citations
An in-depth evaluation of instruction tuning for Arabic NLP tasks using three prominent LLMs: LLaMA 3.1-8B, AceGPT-v2-8B, and Qwen3-8B shows that instruction tuning consistently improves performance across most tasks, with notable variations in effectiveness across different tasks and prompts.
Maged Saeed Al-shaibani, Zaid Alyafeai, Irfan Ahmad· Language Resources and Evalu...· 0 citations
Large Language Models (LLMs) have been rapidly evolving lately, resulting in the need for strong, explainable models to detect the difference between human-generated and machine-generated articles. Existing approaches which are mostly based on fine-tuned transformers suffer from several drawbacks such as rapid obsolescence, paraphrasing attacks, and lack of interpretability. To improve their ability to detect, this paper proposes a novel paradigm called Human vs. LLM Identification (HLI) which introduces a Retrieval-Augmented Generation (RAG)-inspired evidence-based detection strategy alongside a fine-tuned transformer classifier. Our core model, DeBERTa-Sentinel, is built on top of a fine-tuned Microsoft DeBERTa-v3-small model, which uses a disentangled attention mechanism to better capture subtle syntactic and stylistic deviations characteristic of AI-generated text. We evaluate our framework on a balanced dataset of 43,456 text samples, curated from the OpenGPTText corpus and covering AI-generated and human-authored content across diverse domains including news, education, and creative text. The experimental results show improved performance over the selected baselines, with our framework achieving an accuracy of 97.53%, precision of 95.89%, recall of 99.34%, and ROC-AUC of 99.53%. In addition, explainability is integrated into our framework through Local Interpretable Model-agnostic Explanations (LIME) analysis, providing token-level insight into classification decisions. This study establishes a benchmark for scalable, explainable AI text detection, with implications for academic integrity, content moderation, and combating misinformation.
Ibtasam Ur Rehman, Muhammad Islam, Muhammad Yousaf Rehman et al.· Knowledge· 0 citations
This paper tests whether prompting the same (frozen) SLM in three typologically diverse languages and aggregating the outputs can improve classification without retraining or translation, and suggests that cross-lingual diversity rather than surface-level input variation drives the gain.
