Unverified50% confidenceBenchmarkExact time
Reproduction experiments on Qwen3-4B-Base show that MCPO also improves over GRPO, with all three seeds positive and an average improvement of 4.75 points.
1
Sources
50%
Confidence
Long-term
Relevance
11/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUnder the condition of matched per-step generation budgets, FOCUS at the Qwen3.5-9B scale improves by 7.2 percentage points over the strongest baseline (the higher of GRPO/VinePPO) on the BrowseComp-Plus benchmark, reaching 55.4 versus 48.2.65% similarUnverifiedOn the Qwen3.5-2B model, GAD-RL achieved a Micro Recall of 59.92% on CHAOS-Bench, surpassing GRPO and fixed-weight GRPO+OPD by 8.45 and 4.43 percentage points respectively64% similarUnverifiedOn the Qwen3-VL series of models (from 2B to 30B-A3B) and on InternVL3.5-8B, SPLIT-RL improves average accuracy by 1.4 to 6.1 percentage points compared to standard GRPO.64% similarUnverifiedOsprey achieves a 16.1% improvement in mean acceptance length on Qwen3-8B.63% similarUnverifiedThe three-level cache improved generation speed under Q4 weights from 1.83 tokens per second to 8.89, an improvement of nearly 5x.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1157958API
curl https://kongchang.com/api/v1/knowledge/claims/1157958MCP
get_claim(id=1157958)