본문으로 건너뛰기

#Reasoning

75개의 포스트

[논문리뷰] ThinkTwice: Jointly Optimizing Large Language Models for Reasoning and Self-Refinement

본 논문은 Reasoning 최적화와 Self-Refinement 최적화를 하나의 GRPO 프레임워크 안에서 결합한 ThinkTwice를 제안합니다. ThinkTwice는 각 훈련 단계에서 모델이 먼저 Reasoning 문제를 풀고, 동일한 문제에 대해 자신의 이전 답변을 개선하는(Thinking twice) 과정을 연속적으로 수행합니다 .
#Review#Large Language Models#Reinforcement Learning#Reasoning#Self-Refinement#RLVR#Policy Optimization#Implicit Curriculum
댓글 수 로딩 중

[논문리뷰] Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B

arXiv에 게시된 'Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B' 논문에 대한 자세한 리뷰입니다.
#Review#Small Language Models#Reasoning#Diversity Optimization#Supervised Fine-Tuning (SFT)#Reinforcement Learning (RL)#Spectrum-to-Signal Principle (SSP)#Mathematical Reasoning#Code Generation
댓글 수 로딩 중

[논문리뷰] left|,circlearrowright,text{BUS},right|: A Large and Diverse Multimodal Benchmark for evaluating the ability of Vision-Language Models to understand Rebus Puzzles

Deepiha S이 arXiv에 게시한 'left|,circlearrowright,text{BUS},right|: A Large and Diverse Multimodal Benchmark for evaluating the ability of Vision-Language Models to understand Rebus Puzzles' 논문에 대한 자세한 리뷰입니다.
#Review#Vision-Language Models#Multimodal Benchmark#Rebus Puzzles#In-Context Learning#Reasoning#ControlNet#Prompt Engineering
댓글 수 로딩 중

[논문리뷰] R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning

Han Hu이 arXiv에 게시한 'R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning' 논문에 대한 자세한 리뷰입니다.
#Review#Multimodal Large Language Models (MLLMs)#Auto-Thinking#Reinforcement Learning (RL)#Bi-mode Annealing#Bi-mode Policy Optimization (BPO)#General-Purpose AI#Reasoning#Efficiency
댓글 수 로딩 중

[논문리뷰] TreePO: Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling

Zhoufutu Wen이 arXiv에 게시한 'TreePO: Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling' 논문에 대한 자세한 리뷰입니다.
#Review#Reinforcement Learning#Policy Optimization#Large Language Models#Inference Efficiency#Tree Search#Segment-level Decoding#Advantage Estimation#Reasoning
댓글 수 로딩 중

[논문리뷰] On-Policy RL Meets Off-Policy Experts: Harmonizing Supervised Fine-Tuning and Reinforcement Learning via Dynamic Weighting

Guoyin Wang이 arXiv에 게시한 'On-Policy RL Meets Off-Policy Experts: Harmonizing Supervised Fine-Tuning and Reinforcement Learning via Dynamic Weighting' 논문에 대한 자세한 리뷰입니다.
#Review#Large Language Models#Reinforcement Learning#Supervised Fine-Tuning#On-Policy RL#Off-Policy Experts#Dynamic Weighting#LLM Alignment#Reasoning
댓글 수 로딩 중

[논문리뷰] InfiAlign: A Scalable and Sample-Efficient Framework for Aligning LLMs to Enhance Reasoning Capabilities

Zhijie Sang이 arXiv에 게시한 'InfiAlign: A Scalable and Sample-Efficient Framework for Aligning LLMs to Enhance Reasoning Capabilities' 논문에 대한 자세한 리뷰입니다.
#Review#LLM Alignment#Reasoning#Data Curation#Supervised Fine-tuning (SFT)#Direct Preference Optimization (DPO)#Sample Efficiency#Scalability#Multi-dimensional Filtering
댓글 수 로딩 중