When Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?
Modern sequence models are increasingly deployed as agents that interleave token generation with calls to external tools. We give an exact, architecture-level account of when such tool access increases computational expressivity. We model any fixed finite-precision recurrent sequence model, including finite-precision state-space models (SSMs) with $B$ bits of internal state, as a deterministic finite-state controller interacting with an oracle through a finite command/observation interface. Our results form a sharp dichotomy. First, tools that are themselves finite-state add essentially nothin
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.
- PossiblePossibly related (embedding) · 51%phasespace-labs/palinode →
- PossiblePossibly related (embedding) · 48%Hierarchos: Preliminary Findings From a 232M Recurrent Memory-Augmented Assistant Model [P] →
- PossiblePossibly related (embedding) · 46%New Server Hopes to Break Through AI’s “Memory Wall” →
- PossiblePossibly related (embedding) · 50%zircote/rlm-rs →
- FuzzyOverlapping authors or contributors · 62%bytedance/deer-flow →
“Shared author/contributor keys: wang”
- FuzzyOverlapping authors or contributors · 62%ray-project/ray →
“Shared author/contributor keys: wang”
- LinkedLinked via arxiv author · 85%Nikola Zubić →
“When Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?”
- LinkedLinked via arxiv author · 85%Qian Li →
“When Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?”