Metacognitive Behavioral Tuning injects a five-phase structure into LLM reasoning traces to improve accuracy-efficiency on multi-hop QA while reducing trace length and degeneration.
Start by explicitly re-stating the core goal of the problem and filtering out any irrelevant information or distractions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Metacognitive Behavioral Tuning of Large Language Models for Multi-Hop Question Answering
Metacognitive Behavioral Tuning injects a five-phase structure into LLM reasoning traces to improve accuracy-efficiency on multi-hop QA while reducing trace length and degeneration.