Monotonicity Natural Language Inference

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

• Message Passing Inferenceconcept0.756
  Algorithmic framework for probabilistic inference in graphical models.
• Inference-Time Intervention: Eliciting Truthful Answers from a Language Model (Li et al., 2023)concept0.751
  Safety intervention that relies on activation modification, which ESR might undermine
• inference of sentienceconcept0.742
  Attributing subjective experience based on observable embodied behaviours.
• Inference-Time Scalingconcept0.738
  Paradigm of improving model outputs through more computation at inference time; VS is presented as a diversity-oriented alternative
• A diverse set of responses for a conversation captures contradictory ways one could respond, measurable by an NLI modelhypothesis0.738
  Core hypothesis motivating the NLI Diversity metric
• Towards Monosemanticity: Decomposing Language Models with Dictionary Learning (Bricken et al., 2023)concept0.738
  Foundational SAE mechanistic interpretability paper
• Active Inferenceframework0.735
  Foundational framework by Karl Friston; the paper extends it to three hierarchical levels for modeling meta-awareness.
• Friston, FitzGerald et al. (2016) — Active inference and learningconcept0.732
  Prior active inference paper providing detailed neurophysiological implementation of belief updates