method
active
method:hinting-taskHinting Task
One of four ToM tasks analyzed; requires inferring speaker intent from indirect hints; scored 0/1.
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.
- Known cue planted in prompt; ground truth by removing it.
- Neutral hint: suggestion from a person.
- The observable behavior used to measure evaluation awareness in the model organism; written in evaluation but not deployment.
- Neutral hint: prefilled prior model response.
- Neutral hint: answer in XML metadata.
- Neutral hint: correct few-shot answers marked with symbols.
- The problem of ensuring all tasks in MTL perform well, avoiding dominance by some tasks.
- Task where a random word is prefilled as the assistant's response, then the model is asked whether it intended to say that word, testing introspection on prior intentions.