AI Hub

Research Papers

The research that matters, distilled — search, filter by topic, sort, and group.

107 results

Topics
gpt-oss-120b & gpt-oss-20b Model CardOpenAI
ArchitectureTrainingSafety
Aug 11, 2025
GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation ModelsZ.ai (Zhipu AI)
ArchitectureTrainingReinforcement Learning
Aug 8, 2025
Kimi K2: Open Agentic IntelligenceMoonshot AI
ArchitectureTrainingReinforcement Learning
Jul 28, 2025
Group Sequence Policy OptimizationAlibaba (Qwen Team)
Reinforcement LearningTraining
Jul 24, 2025
Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level ComputationKAIST AI / Google DeepMind / Mila
Architecture
Jul 14, 2025
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic CapabilitiesGoogle DeepMind
ArchitectureTrainingEvaluation
Jul 8, 2025
MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning AttentionMiniMax AI
ArchitectureReinforcement LearningEvaluation
Jun 16, 2025
Qwen3 Technical ReportAlibaba (Qwen Team)
ArchitectureTrainingReinforcement Learning
May 14, 2025
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement LearningDeepSeek
Reinforcement LearningTraining
Jan 22, 2025
DeepSeek-V3 Technical ReportDeepSeek
ArchitectureTraining
Dec 27, 2024
The Llama 3 Herd of ModelsMeta
ArchitectureTraining
Jul 31, 2024
Qwen2 Technical ReportAlibaba
Training
Jul 15, 2024
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts ModelDeepSeek
ArchitectureTraining
May 7, 2024
Phi-3 Technical ReportMicrosoft
Training
Apr 22, 2024
Gemma: Open Models Based on Gemini Research and TechnologyGoogle
Training
Mar 13, 2024
Gemini 1.5: Unlocking Multimodal Understanding Across Millions of Tokens of ContextGoogle
ArchitectureEvaluation
Mar 8, 2024
Mixtral of ExpertsMistral AI
Architecture
Jan 8, 2024
Gemini: A Family of Highly Capable Multimodal ModelsGoogle
ArchitectureEvaluation
Dec 19, 2023
Mamba: Linear-Time Sequence Modeling with Selective State SpacesCarnegie Mellon
Architecture
Dec 1, 2023
Mistral 7BMistral AI
Architecture
Oct 10, 2023
Llama 2: Open Foundation and Fine-Tuned Chat ModelsMeta
TrainingSafety
Jul 18, 2023
Textbooks Are All You NeedMicrosoft
Training
Jun 20, 2023
Direct Preference Optimization: Your Language Model is Secretly a Reward ModelStanford
Reinforcement LearningTraining
May 29, 2023
QLoRA: Efficient Finetuning of Quantized LLMsUniversity of Washington
Training
May 23, 2023