Unverified50% confidenceBenchmarkExact time
Under the single-step inference setting, BCSM achieves a 15.1% relative improvement on MMMU.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/4/2026
First Seen
Valid until: 2/2/2027
Sources
Related Entities
Related Claims
UnverifiedMMEA improves over baselines by 28.7 percentage points on mental state and behavior prediction tasks.67% similarUnverifiedUnder the single-step inference setting, BCSM reduces the number of neural function evaluations (NFE) by up to 6 times compared with baselines, while achieving a 16.9% relative improvement on HumanEval.64% similarUnverifiedIn benchmark testing, token usage surged by 1500% (approximately 15x) going from x-high to Max mode.62% similarUnverified多角色流水线（Self-Refine）的输出质量通常比单次LLM调用提升15%-30%61% similarUnverifiedExperiments show that IDM improves Recall@10 by an average of 15.39%, and CMM improves NDCG@10 by an average of 8.28%.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1133925API
curl https://kongchang.com/api/v1/knowledge/claims/1133925MCP
get_claim(id=1133925)