Inference-Time Scaling

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

• Inference-Time Steeringconcept0.816
  Subtracting a scaled persona vector from hidden states at each decoding step to reduce trait expression after finetuning
• Inference-Time Controlconcept0.814
  Approach to steering LLM behavior without modifying weights, offering flexibility and composability
• Test-Time Scalingconcept0.810
  Approach using extra compute at test time to double-check answers and improve reliability.
• Inference-Time Intervention: Eliciting Truthful Answers from a Language Model (Li et al., 2023)concept0.795
  Safety intervention that relies on activation modification, which ESR might undermine
• Inference-Time Intervention (ITI)method0.784
  Method by Li et al. 2023a that adds static vectors to model activations at inference time to steer behavior
• Message Passing Inferenceconcept0.777
  Algorithmic framework for probabilistic inference in graphical models.
• Stages of Inferenceframework0.776
  The perspective that LLM inference decomposes into distinct computational stages, which the paper extends to looped models
• Inference costconcept0.770
  Computational expense proportional to number of generated tokens, targeted for reduction by ReflCtrl