Metacognitive Consolidation lets LLMs accumulate reusable meta-reasoning skills from past episodes to improve future performance across benchmarks.
Title resolution pending
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.
Macro uses Direct Preference Optimization on composite-scored preference pairs to improve validity of multilingual self-generated counterfactual explanations by 12.55% on average without degrading minimality.
RPSFT improves the in-domain versus out-of-domain performance trade-off during LLM supervised fine-tuning by penalizing rotations in pretrained singular subspaces as a proxy for loss-sensitive directions.
Math reasoning gains in LLMs rarely transfer to general domains; RL tuning generalizes while SFT causes forgetting and representation drift.
LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.
citing papers explorer
-
Beyond Meta-Reasoning: Metacognitive Consolidation for Self-Improving LLM Reasoning
Metacognitive Consolidation lets LLMs accumulate reusable meta-reasoning skills from past episodes to improve future performance across benchmarks.
-
SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution
SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.
-
Enhancing Multilingual Counterfactual Generation through Alignment-as-Preference Optimization
Macro uses Direct Preference Optimization on composite-scored preference pairs to improve validity of multilingual self-generated counterfactual explanations by 12.55% on average without degrading minimality.
-
Rotation-Preserving Supervised Fine-Tuning
RPSFT improves the in-domain versus out-of-domain performance trade-off during LLM supervised fine-tuning by penalizing rotations in pretrained singular subspaces as a proxy for loss-sensitive directions.
-
Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning
Math reasoning gains in LLMs rarely transfer to general domains; RL tuning generalizes while SFT causes forgetting and representation drift.
-
LIMO: Less is More for Reasoning
LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.
- Simply Stabilizing the Loop via Fully Looped Transformer