Encoded but Not Actionable: Auditing the Decode-Generate-Steer Gap in Frozen LLMs for Geometric Constraints
Large language models (LLMs) have demonstrated strong performance on structured reasoning tasks, but what they encode and whether it informs model behavior remain unclear. We investigate this question through geometric reasoning, using parametric CAD constraints as a controlled testbed for separating local pairwise relations from sketch-level constraint status. By probing the hidden states of six frozen decoder-only LLMs, we examine four properties: linear decodability, forced-choice generation, activation-level influence, and behavioral steerability. Pretraining substantially improves the dec
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzyOverlapping authors or contributors · 62%DietrichGebert/ponytail →
“Shared author/contributor keys: cheng”
- LinkedLinked via arxiv author · 85%Xiaoman Liang →
“Encoded but Not Actionable: Auditing the Decode-Generate-Steer Gap in Frozen LLMs for Geometric Constraints”
- LinkedLinked via arxiv author · 85%Xinzhao Cheng →
“Encoded but Not Actionable: Auditing the Decode-Generate-Steer Gap in Frozen LLMs for Geometric Constraints”
- LinkedLinked via arxiv author · 85%Faizan Wajid →
“Encoded but Not Actionable: Auditing the Decode-Generate-Steer Gap in Frozen LLMs for Geometric Constraints”