LLM reasoning / chain-of-thought

Mixture-of-Agents and LLM-Blender (runtime ensembles)

Q: Is Mixture-of-Agents and LLM-Blender (runtime ensembles) superseded?

Mixture-of-Agents and LLM-Blender (runtime ensembles) (LLM reasoning / chain-of-thought): cited as a baseline — critiqued by newer work, not yet beaten on a benchmark here. 1 paper(s) critique it, 0 beat it on benchmarks — #457 of 772 most-superseded. Sub-problem: cluster led by single-teacher CoT distillation. Newer alternatives in the same sub-problem include CoT-Evo.

Superseded baseline#457 of 772 most-superseded

Cited as a baseline — critiqued by newer work, not yet beaten on a benchmark here

1 papers critique it · 0 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites Mixture-of-Agents and LLM-Blender (runtime ensembles) as a baseline.

While effective, they inevitably introduce significant inference overhead and high Time to First Token (TTFT), prohibiting deployment in resource-constrained environments.

— "The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.

CoT-EvoCoT-Evo: Evolutionary Distillation of Chain-of-Thought for Scientific Reasoning
Oct 15, 2025