Extracting Small Translation Specialists from LLMs by Aggressively Pruning Experts

arXiv:2605.28042 · cs.CL, cs.AI, cs.LG · Submitted 2026-05-27 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2026-05-27

Updated: 2026-08-31

Terminology

• Pruning — 81× in this paper · explained in 4 episodes
• Calibration — 38× in this paper · explained in 7 episodes
• Core — 24× in this paper · explained in 2 episodes
• Ablation — 12× in this paper · explained in 1 episode
• Generalization — 10× in this paper · explained in 13 episodes
• Distillation — 8× in this paper · explained in 2 episodes
• Fine-Tuning — 6× in this paper · explained in 5 episodes
• Decode — 4× in this paper · explained in 2 episodes
• Interpretability — 3× in this paper · explained in 15 episodes
• Large Language Models (LLMs) — 3× in this paper · explained in 6 episodes
• Modularity — 3× in this paper · explained in 9 episodes
• Sparsity — 3× in this paper · explained in 4 episodes
• Transformer — 3× in this paper · explained in 2 episodes
• Benchmark — 2× in this paper · explained in 2 episodes
• Mixture of Experts — 2× in this paper · explained in 2 episodes
• Quantization — 2× in this paper · explained in 12 episodes
• Reconstruction Error — 2× in this paper · explained in 2 episodes
• Redundancy — 2× in this paper · explained in 2 episodes
• Robustness — 2× in this paper · explained in 5 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Generalizability — 1× in this paper · explained in 6 episodes
• Model Merging — 1× in this paper · explained in 2 episodes
• Modular Components — 1× in this paper · explained in 2 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes
• Resource Constraints — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes

Sources

• Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency
• Translation as a Scalable Proxy for Multilingual Evaluation
• TranslateGemma Technical Report
• BanglaSTEM: A Parallel Corpus for Technical Domain Bangla-English Translation
• Distilling the Knowledge in a Neural Network
• What do Language Models Learn and When? The Implicit Curriculum Hypothesis
• No Language Left Behind: Scaling Human-Centered Machine Translation
• gpt-oss-120b & gpt-oss-20b Model Card
• Omnilingual MT: Machine Translation for 1,600 Languages
• EMO: Pretraining Mixture of Experts for Emergent Modularity
• OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
• Qwen3 Technical Report
• Does a Global Perspective Help Prune Sparse MoEs Elegantly?
• Hunyuan-MT Technical Report

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering