Unverified50% confidenceBenchmarkExact time
Under the two-reward setting, WAS-PO's training reward is lower than GDPO's.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/3/2026
First Seen
Valid until: 2/1/2027
Sources
Related Entities
Related Claims
UnverifiedAmong the three DPO attempts, one reached a reward margin of 1.0 during training, but its free-generation outputs showed no change at all.69% similarVerifiedCompared to PPO, GRPO eliminates the value network (Critic) and directly compares rewards among samples within a group to estimate the advantage function, significantly reducing training computation costs.66% similarUnverifiedpenalty_joint_pos_init gradually reduces its weight as training progresses through curriculum learning, and the penalty_slip anti-slip penalty has not yet been implemented.65% similarUnverifiedGRPO相比PPO降低了约50%的训练显存需求63% similarUnverifiedThe paper finds that the ranking of downstream tasks is not consistent with the ranking of training loss, and schemes with lower training loss do not necessarily perform better in downstream evaluations.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1130539API
curl https://kongchang.com/api/v1/knowledge/claims/1130539MCP
get_claim(id=1130539)