Enhancing Low-Resource Language Reasoning via High-Resource Language Feature Transfer

arXiv:2608.30462 · cs.CL, cs.AI · Submitted 2026-08-31 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-08-31

Updated: 2026-08-31

Code: https://github.com/decoderesearch/SAELens

Terminology

• Steering — 72× in this paper · explained in 2 episodes
• Accuracy — 20× in this paper · explained in 1 episode
• Benchmark — 20× in this paper · explained in 2 episodes
• Trace — 18× in this paper · explained in 2 episodes
• Transformer — 13× in this paper · explained in 2 episodes
• Low-Resource Language — 8× in this paper
• Reasoning Traces — 8× in this paper · explained in 3 episodes
• Ablation — 7× in this paper · explained in 1 episode
• Adam — 6× in this paper · explained in 2 episodes
• Interpretability — 5× in this paper · explained in 15 episodes
• Feature Selection — 4× in this paper · explained in 4 episodes
• High-Resource Language — 4× in this paper
• Low-Resource Languages — 4× in this paper · explained in 4 episodes
• Activation Addition — 3× in this paper · explained in 1 episode
• Generalization — 3× in this paper · explained in 13 episodes
• Aggregation — 2× in this paper · explained in 1 episode
• Consistency — 2× in this paper · explained in 3 episodes
• Contrastive Activation Addition — 2× in this paper
• Instruction Tuning — 2× in this paper · explained in 2 episodes
• Mechanistic interpretability — 2× in this paper · explained in 2 episodes
• Representation Engineering — 2× in this paper · explained in 2 episodes
• Sparse Autoencoders (SAEs) — 2× in this paper · explained in 3 episodes
• Activation Patching — 1× in this paper · explained in 2 episodes
• Activation Steering — 1× in this paper · explained in 4 episodes
• Chain-of-Thought (CoT) — 1× in this paper · explained in 4 episodes
• Chain-of-Thought (CoT) Reasoning — 1× in this paper · explained in 2 episodes
• Copula — 1× in this paper · explained in 2 episodes
• Cross-lingual Transfer — 1× in this paper · explained in 2 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Feature Extraction — 1× in this paper · explained in 2 episodes
• Fidelity — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Hallucination — 1× in this paper · explained in 4 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Latent Representation — 1× in this paper · explained in 2 episodes
• Latent Space — 1× in this paper · explained in 7 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Tokenization — 1× in this paper · explained in 2 episodes
• Transformer Models — 1× in this paper · explained in 2 episodes

Sources

• Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization
• Improving Steering Vectors by Targeting Sparse Autoencoder Features
• Sparse Autoencoders Find Highly Interpretable Features in Language Models
• Toy Models of Superposition
• Gemma 2: Improving Open Language Models at a Practical Size
• Let's Verify Step by Step
• MMATH: A Multilingual Benchmark for Mathematical Reasoning
• Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
• CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages
• Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models?
• GlotLID: Language Identification for Low-Resource Languages
• Steering Llama 2 via Contrastive Activation Addition
• Programming Refusal with Conditional Activation Steering
• Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2
• Qwen2.5 Technical Report
• Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders
• Language Models are Multilingual Chain-of-Thought Reasoners
• Attribution Patching Outperforms Automated Circuit Discovery
• Steering Language Models With Activation Engineering
• HuggingFace's Transformers: State-of-the-art Natural Language Processing

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