GeoMin: Data-Efficient Semi-Supervised RLVR via Geometric Distribution Modeling
Reinforcement learning with verifiable rewards (RLVR) significantly advances LLM reasoning, yet it faces a dilemma: standard supervised scaling is throttled by high annotation costs, while unsupervised alternatives suffer from severe model collapse. Recent semi-supervised RLVR methods address this by using a small labeled set to guide unlabeled data, achieving a promising trade-off between training efficacy and annotation cost. However, they suffer from a severe data-efficiency bottleneck due to the reliance on coarse performance heuristics, leaving a vast majority of valuable instances underutilized. To this end, we propose GeoMin, which models global feature distributions on labeled data to decode the structural discrepancy between correct and incorrect rollouts, thereby establishing a robust prior to assess the reliability of self-reward signals and fully unleash the potential of unlabeled data. Empirically, GeoMin outperforms the strongest baselines by +4.1% and even surpasses fully supervised models with only 10% of the annotations, demonstrating remarkable data efficiency.
Code (0)
등록된 구현이 없습니다.
Tasks
Reinforcement LearningSimilar Papers 제목 키워드 기반
Detecting Events in Crowds Through Changes in Geometrical Dimensions of Pedestrians
Security is an important topic in our contemporary world, and the ability to automate the detection of any events of interest that can take place in a crowd is of great interest to a population. We hypothesize that the d…
validGeoMind: An Agentic Workflow for Lithology Classification with Reasoned Tool Invocation
Lithology classification in well logs is a fundamental geoscience data mining task that aims to infer rock types from multi dimensional geophysical sequences. Despite recent progress, existing approaches typically formul…
GeoRA: Geometry-Aware Low-Rank Adaptation for RLVR
Reinforcement Learning with Verifiable Rewards (RLVR) is a key paradigm for improving large-scale reasoning models. Unlike supervised fine-tuning (SFT), RLVR exhibits distinct optimization dynamics and is sensitive to th…
Reinforcement LearningTraPO: A Semi-Supervised Reinforcement Learning Framework for Boosting LLM Reasoning
Reinforcement learning with verifiable rewards (RLVR) has proven effective in training large reasoning models (LRMs) by leveraging answer-verifiable signals to guide policy optimization, which, however, suffers from high…
Reinforcement LearningMathematical ReasoningMasked-and-Reordered Self-Supervision for Reinforcement Learning from Verifiable Rewards
Test-time scaling has been shown to substantially improve large language models' (LLMs) mathematical reasoning. However, for a large portion of mathematical corpora, especially theorem proving, RLVR's scalability is limi…
Reinforcement LearningMathematical Reasoning