Identifying Crucial Attention Heads for Multilingual Language Models: Retrieval and Retrieval-Transition Heads

arXiv:2602.22453 · cs.CL · Submitted 2026-02-25 · Read on arXiv

cs.CL

Submitted: 2026-02-25

Updated: 2026-09-11

Terminology

• Retrieval Heads — 37× in this paper
• Retrieval-Transition Heads — 26× in this paper
• Benchmark — 12× in this paper · explained in 2 episodes
• Failure Modes — 9× in this paper · explained in 3 episodes
• Transformer — 8× in this paper · explained in 2 episodes
• Accuracy — 4× in this paper · explained in 1 episode
• Latent Reasoning — 4× in this paper · explained in 2 episodes
• Loss of Coherence — 4× in this paper
• Low-Resource Languages — 4× in this paper · explained in 4 episodes
• Hallucination — 3× in this paper · explained in 3 episodes
• In-Context Learning — 3× in this paper · explained in 4 episodes
• KV cache — 3× in this paper · explained in 3 episodes
• Latent Space — 3× in this paper · explained in 7 episodes
• Retrieval Transition Score — 3× in this paper
• Sparsity — 3× in this paper · explained in 4 episodes
• Alignment — 2× in this paper · explained in 1 episode
• F1 Score — 2× in this paper · explained in 3 episodes
• LLM-as-a-Judge — 2× in this paper · explained in 3 episodes
• Pruning — 2× in this paper · explained in 4 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Chain-of-Thought (CoT) — 1× in this paper · explained in 4 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Core — 1× in this paper · explained in 2 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Transferability — 1× in this paper · explained in 4 episodes
• Transformer Architecture — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes
• Zero-Shot Transfer — 1× in this paper · explained in 4 episodes

Sources

• Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
• The Emergence of Abstract Thought in Large Language Models Beyond Any Language
• Focusing on Language: Revealing and Exploiting Language Attention Heads in Multilingual Large Language Models
• Training Verifiers to Solve Math Word Problems
• DeepSeek-V3 Technical Report
• Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
• Qwen2.5 Technical Report
• Language Models are Multilingual Chain-of-Thought Reasoners
• RazorAttention: Efficient KV Cache Compression Through Retrieval Heads
• Lost in Multilinguality: Dissecting Cross-lingual Factual Inconsistency in Transformer Language Models
• The Llama 3 Herd of Models
• Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced Optimization
• Retrieval Head Mechanistically Explains Long-Context Factuality
• DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
• Qwen3 Technical Report
• Babel: Open Multilingual Large Language Models Serving Over 90% of Global Speakers
• Attention Heads of Large Language Models: A Survey

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