RIFT: Token Influence Follows the Path of Least Resistance

cs.CL, cs.LG

Submitted: 2026-10-12

Updated: 2026-10-12

Comments: 9 pages, 5 figures, 4 tables

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

Sources

• AtP*: An efficient and scalable method for localizing LLM behaviour to components
• Learning to Attribute with Attention
• Lost in the Middle at Birth: An Exact Theory of Transformer Position Bias
• A Structural Theory of Position Bias in Transformers
• Attention Flows for General Transformers
• In-context Learning and Induction Heads
• Understanding the Failure Modes of Transformers through the Lens of Graph Neural Networks
• Qwen3 Technical Report
• The Llama 3 Herd of Models

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