DeepPrune: Parallel Scaling without Inter-trace Redundancy Paper β’ 2510.08483 β’ Published Oct 9, 2025 β’ 24 β’ 2
MMR-V: What's Left Unsaid? A Benchmark for Multimodal Deep Reasoning in Videos Paper β’ 2506.04141 β’ Published Jun 4, 2025 β’ 31 β’ 2
Establishing Trustworthy LLM Evaluation via Shortcut Neuron Analysis Paper β’ 2506.04142 β’ Published Jun 4, 2025 β’ 28 β’ 2
LongWriter-V: Enabling Ultra-Long and High-Fidelity Generation in Vision-Language Models Paper β’ 2502.14834 β’ Published Feb 20, 2025 β’ 24 β’ 2