Target-Language Generation in Multilingual Models: Activation Steering and Optimal Control

arXiv:2609.16967 · cs.CL · Submitted 2026-09-15 · Read on arXiv

cs.CL

Submitted: 2026-09-15

Updated: 2026-09-15

Comments: Accepted at EMNLP 2026

Code: https://github.com/chengemily1/llm-control

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

Terminology

• Steering — 42× in this paper · explained in 2 episodes
• Activation Steering — 25× in this paper · explained in 4 episodes
• Perplexity — 6× in this paper · explained in 3 episodes
• Transformer — 6× in this paper · explained in 2 episodes
• Accuracy — 5× in this paper · explained in 1 episode
• Alignment — 5× in this paper · explained in 1 episode
• Benchmark — 5× in this paper · explained in 2 episodes
• Generalization — 4× in this paper · explained in 13 episodes
• Adam — 3× in this paper · explained in 2 episodes
• Embeddings — 3× in this paper · explained in 2 episodes
• Latent Space — 3× in this paper · explained in 7 episodes
• Representation Learning — 3× in this paper · explained in 2 episodes
• Activation Addition — 2× in this paper · explained in 1 episode
• Consistency — 2× in this paper · explained in 3 episodes
• In-Context Learning — 2× in this paper · explained in 4 episodes
• Theoretical Guarantees — 2× in this paper · explained in 2 episodes
• Activation Patching — 1× in this paper · explained in 2 episodes
• Anisotropy — 1× in this paper · explained in 3 episodes
• Core — 1× in this paper · explained in 2 episodes
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Edge AI — 1× in this paper · explained in 2 episodes
• Few-Shot Learning — 1× in this paper · explained in 4 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Latent Representation — 1× in this paper · explained in 2 episodes
• Linear probe — 1× in this paper · explained in 2 episodes
• Low-Resource Languages — 1× in this paper · explained in 4 episodes
• Meta-learning — 1× in this paper · explained in 2 episodes
• Prompt Adherence — 1× in this paper · explained in 2 episodes
• Test-time Scaling — 1× in this paper · explained in 2 episodes
• alpha — 1× in this paper · explained in 1 episode

Sources

• BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
• When Is Multilinguality a Curse? Language Modeling for 250 High- and Low-Resource Languages
• No Language Left Behind: Scaling Human-Centered Machine Translation
• Gemma 3 Technical Report
• ECLeKTic: a Novel Challenge Set for Evaluation of Cross-Lingual Knowledge Transfer
• The Llama 3 Herd of Models
• In-context Learning and Induction Heads
• Qwen2.5 Technical Report
• Cross-Lingual Generalization and Compression: From Language-Specific to Shared Neurons
• Tiny Aya: Bridging Scale and Multilingual Depth
• Extracting Latent Steering Vectors from Pretrained Language Models
• Steering Language Models With Activation Engineering
• Lost in Multilinguality: Dissecting Cross-lingual Factual Inconsistency in Transformer Language Models
• Qwen3 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