Lost in Reordering: Structural Sensitivity of Multilingual LLMs under Semantics-Preserving Perturbations

arXiv:2609.03511 · cs.CL · Submitted 2026-09-03 · Read on arXiv

cs.CL

Submitted: 2026-09-03

Updated: 2026-09-03

Comments: Accepted at EMNLP 2026 (Findings - Long paper)

Code: https://github.com/karthika95/lost-in-reordering

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Activation Patching — 17× in this paper · explained in 2 episodes
• Benchmark — 14× in this paper · explained in 2 episodes
• Robustness — 13× in this paper · explained in 5 episodes
• Accuracy — 10× in this paper · explained in 1 episode
• Interpretability — 8× in this paper · explained in 15 episodes
• Adaptation — 5× in this paper · explained in 1 episode
• Alignment — 5× in this paper · explained in 1 episode
• Mechanistic interpretability — 4× in this paper · explained in 2 episodes
• Transformer — 4× in this paper · explained in 2 episodes
• Benchmarking — 3× in this paper · explained in 4 episodes
• Large Language Models (LLMs) — 3× in this paper · explained in 6 episodes
• Cosine Similarity — 2× in this paper · explained in 3 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Verification — 2× in this paper · explained in 2 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Core — 1× in this paper · explained in 2 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Hallucination — 1× in this paper · explained in 4 episodes
• Semantic Similarity — 1× in this paper · explained in 2 episodes
• Structural Constraints — 1× in this paper · explained in 2 episodes
• Thinking mode — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• gpt-oss-120b & gpt-oss-20b Model Card
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Training Verifiers to Solve Math Word Problems
• The Llama 3 Herd of Models
• MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations
• Benchmarking Hindi LLMs: A New Suite of Datasets and a Comparative Analysis
• Semantically Cohesive Word Grouping in Indian Languages
• Same Meaning, Different Scores: Lexical and Syntactic Sensitivity in LLM Evaluation
• A Three-Pronged Approach to Cross-Lingual Adaptation with Multilingual LLMs
• Gemma 4 Technical Report
• Gemma 2: Improving Open Language Models at a Practical Size
• Mathematics Isn't Culture-Free: Probing Cultural Gaps via Entity and Scenario Perturbations
• Qwen3 Technical Report
• Exploring the Limitations of Large Language Models in Compositional Relation Reasoning

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