Unverified50% confidenceBenchmarkExact time
On Sudoku and maze tasks, CRAM reduces inference FLOPs by at least 282 times compared with recurrent models.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/13/2026
First Seen
Valid until: 2/11/2027
Sources
Related Entities
Related Claims
UnverifiedCompared with direct continuous-time sampling, SpecCTMC reduces the number of sequential model evaluations by 1.73 to 4.38 times while maintaining the reference sampling distribution.66% similarUnverifiedIn calibrated anisotropic Gaussian simulations, DI-SCUSUM performs close to likelihood-ratio CUSUM while reducing detection delay by about 91% compared with score-based CUSUM (SCUSUM).64% similarUnverified将FP32权重降至INT8后,典型Transformer模型基准测试分数下降通常不超过1%,推理速度提升2至4倍,内存占用减少75%63% similarUnverifiedFew-shot reasoning significantly reduced the hallucination rate: the strongest model produced unsupported QoS requirements in about 1% of responses, whereas performance under the zero-shot setting was markedly worse.61% similarUnverified研究表明当ReAct循环超过10步时,模型的任务完成率会出现显著下降60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1166107API
curl https://kongchang.com/api/v1/knowledge/claims/1166107MCP
get_claim(id=1166107)