COMPACT adaptively fuses multi-teacher CoT supervisions using graph-based consensus, mutual-information adaptability, and loss-based difficulty metrics to improve small language model reasoning performance while mitigating catastrophic forgetting.
The ba- sic statistics of these benchmarks are presented in Tables 2–5
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
"The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework
COMPACT adaptively fuses multi-teacher CoT supervisions using graph-based consensus, mutual-information adaptability, and loss-based difficulty metrics to improve small language model reasoning performance while mitigating catastrophic forgetting.