Adapting Qwen3 to Tajik from 0.6B to 8B: Lower Text Loss Did Not Mean Better Understanding
cs.CL
Submitted: 2026-11-04
Updated: 2026-11-04
Terminology
Sources
- Benchmarking Parameter-Efficient Fine-Tuning of Large Language Models for Low-Resource Tajik Text Generation with the Tajik Web Corpus
- TajikNLP: An Open-Source Toolkit for Comprehensive Text Processing of Tajik (Cyrillic Script)
- LoRA Learns Less and Forgets Less
- When Is Multilinguality a Curse? Language Modeling for 250 High- and Low-Resource Languages
- Efficient and Effective Text Encoding for Chinese LLaMA and Alpaca
- QLoRA: Efficient Finetuning of Quantized LLMs
- Continual Pre-Training of Large Language Models: How to (re)warm your model?
- LoRA: Low-Rank Adaptation of Large Language Models
- Simple and Scalable Strategies to Continually Pre-train Large Language Models
- MADLAD-400: A Multilingual And Document-Level Large Audited Dataset
- An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning
- SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights
- FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language
- No Language Left Behind: Scaling Human-Centered Machine Translation
- Qwen3 Technical Report
- Large Language Models Are Not Robust Multiple Choice Selectors
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data