ODAR: Principled Adaptive Routing for LLM Reasoning via Active Inference
arXiv:2602.23681v1 Announce Type: new Abstract: The paradigm of large language model LLM reasoning is shifting from parameter scaling to testtime compute scaling, yet many existing approaches still rely on uniform bruteforce sampling for example, fixed bestofN or selfconsistency that is costly,...
arXiv:2602.23681v1 Announce Type: new Abstract: The paradigm of large language model LLM reasoning is shifting from parameter scaling to testtime compute scaling, yet many existing approaches still rely on uniform bruteforce sampling for example, fixed bestofN or selfconsistency that is costly, hard to attribute, and can trigger overthinking with diminishing returns.
This content is a summary of the original article from ArXiv AI. Please visit the original site for the full article.
Read Original Article →