Geometry of Divergence: Tracking Hidden-State Trajectories for Adaptive Multi-Turn Reasoning
尚未评估计划受阻指标结果claim-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.
来源：paper:Page 6, Section 6, Bullet 'Action-conditioned Policies'
报告指标与观测值
f1_score
meas-f1-read
论文报告 0.826 fraction
实际观测 — fraction
f1_score
meas-f1-respond
论文报告 0.711 fraction
实际观测 — fraction
f1_score
meas-f1-write
论文报告 0.7 fraction
实际观测 — fraction
f1_score
meas-f1-transfer
论文报告 0.531 fraction
实际观测 — fraction
Assessment（0）
这条 Claim 暂无已发布的不可变 Assessment。
关联运行（0）
这条 Claim 暂无关联执行。