framework
active
framework:ar-modelAR(ω) model
Stochastic process model predicting next token from a context window of length ω; mapped to local Hamiltonian
Neighborhood — ranked by edge-count
Thinkers (1)
thinker
- George E P BoxintroducesCo-author of Box & Jenkins (1970), foundational work on autoregressive models
Concepts (2)
concept
- Local Hamiltonianassociated_withSum of windowed Hamiltonians with same window length; a key construct introduced in the paper to model local interactions on graphs
- context windowassociated_withFinite number of previous tokens used by autoregressive models to predict the next token; defines interaction range
Findings (1)
finding
- Autoregressive model unable to converge to a single stored pattern for any finite β (Corollary 2)aboutConsequence of Theorem 3 and 1D no-order result
Frameworks (1)
framework
- transformer architectureimplementsNeural network architecture based on attention, commonly used in large language models
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Statistical technique where outputs are regressed on previous values; used in language generation
- Second model system studied; used to show why flat autoregressive LLMs struggle with long-range coherence.
- A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
- A message-passing concurrency model where processes (actors) communicate via messages (talks) and generate new processes; related to concurrent objects.
- A 1:50 scale model used for overall design simulation of the Athens Megaron spaces and floors.
- Concurrency control using monitors with condition queues; compared to Linda's tuple space.