CLIMB: CLustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training
Paper • 2504.13161 • Published • 98
stuff i never have time to read
-
-
Hebbian Learning based Orthogonal Projection for Continual Learning of Spiking Neural Networks
Paper • 2402.11984 • Published -
BlackGoose Rimer: Harnessing RWKV-7 as a Simple yet Superior Replacement for Transformers in Large-Scale Time Series Modeling
Paper • 2503.06121 • Published • 5 -
Timer: Transformers for Time Series Analysis at Scale
Paper • 2402.02368 • Published • 2 -
Timer-XL: Long-Context Transformers for Unified Time Series Forecasting
Paper • 2410.04803 • Published • 2 -
Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts
Paper • 2409.16040 • Published • 16 -
Packing Input Frame Context in Next-Frame Prediction Models for Video Generation
Paper • 2504.12626 • Published • 51 -
One RL to See Them All: Visual Triple Unified Reinforcement Learning
Paper • 2505.18129 • Published • 63 -
Ming-Omni: A Unified Multimodal Model for Perception and Generation
Paper • 2506.09344 • Published • 33 -
APE: Faster and Longer Context-Augmented Generation via Adaptive Parallel Encoding
Paper • 2502.05431 • Published • 6 -
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Paper • 2507.06261 • Published • 68 -
UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
Paper • 2509.02544 • Published • 128 -
Qwen3-ASR Technical Report
Paper • 2601.21337 • Published • 38 -
Mercury: Ultra-Fast Language Models Based on Diffusion
Paper • 2506.17298 • Published • 11