Geometry of Divergence: Tracking Hidden-State Trajectories for Adaptive Multi-Turn Reasoning
Not assessedPlan blockedMeasurementclaim-action-classifier-f1
The learned classifier predicting the agent's next action achieves F1 scores of 0.826 for Read, 0.711 for Respond, 0.700 for Write, and 0.531 for Transfer.
Source: paper:Page 6, Section 6, Bullet 'Action-conditioned Policies'
Reported and observed measurements
f1_score
meas-f1-read
Reported 0.826 fraction
Observed — fraction
f1_score
meas-f1-respond
Reported 0.711 fraction
Observed — fraction
f1_score
meas-f1-write
Reported 0.7 fraction
Observed — fraction
f1_score
meas-f1-transfer
Reported 0.531 fraction
Observed — fraction
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.