concept
active
concept:multinomial-sampling-noise-model

Multinomial sampling noise model

Counts c_t ~ Multinomial(S,o_t).

Neighborhood — ranked by edge-count

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

  • The mechanism by which LLMs generate text: drawing a token from the next-token distribution and appending it to context repeatedly
  • Naive Samplingmethod0.733
    Baseline using Binomial(k,p) draws from precomputed rollouts.
  • Test-time compute scaling guided by reward probes with majority-vote over action.
  • Claude 4.5 Haiku used to segment responses into attempts and score each attempt 0-100 for relevance
  • Technique used to demonstrate that the self-prior captures visual–proprioceptive associations by recovering visual appearance from proprioception alone
  • Autoregressive modelsframework0.719
    Second model system studied; used to show why flat autoregressive LLMs struggle with long-range coherence.
  • The joint distribution over events in the world that generate observed data; the proposed endpoint of representational convergence
  • Statistical technique where outputs are regressed on previous values; used in language generation