Scholar Feed

LLM reasoning / chain-of-thought

single-teacher CoT distillation

Superseded baseline#94 of 772 most-superseded

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

2 papers critique it · 0 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites single-teacher CoT distillation as a baseline.

Traditional approaches force student SLMs to fit a single teacher-generated rationale, neglecting the diversity in reasoning paths that could lead to the SLM simply mimics the superficial style of the teacher model's outputs
— "The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework
“single-teacher optimization may introduce bias”
— CoT-Evo: Evolutionary Distillation of Chain-of-Thought for Scientific Reasoning

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Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.