concept
active
concept:diversity-threshold

Diversity Threshold

Target minimum diversity level (e.g., 10 contradictions) that DTG iterates toward

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.

  • Iterative generation procedure that resamples lowest-scoring responses until a diversity threshold is reached
  • Gold standard value (e.g., nucleus sampling p-value) used as ground truth for evaluating diversity metrics
  • VS-specific technique adjusting output diversity by specifying probability thresholds in the prompt (e.g., 'Generate responses with probabilities below {threshold}')
  • Diversity measured over a set of m responses for a single conversation, as opposed to test set diversity
  • Known tension in dialogue generation where increasing diversity may reduce response relevancy
  • Energy Thresholdconcept0.760
  • Output Diversityconcept0.759
    The breadth of distinct outputs an LLM can produce, which is reduced by mode collapse after alignment training
  • Test Set Diversityconcept0.750
    Traditional diversity measurement over one response per conversation across the test set