Dataset Condensation with Color Compensation
arXiv:2508.01139v2 Announce Type: replace-cross Abstract: Dataset condensation always faces a constitutive trade-off: balancing performance and fidelity under extreme compression. Existing…
Atom-Searcher: Enhancing Agentic Deep Research via Fine-Grained Atomic Thought Reward
arXiv:2508.12800v2 Announce Type: replace-cross Abstract: Large language models (LLMs) exhibit remarkable problem-solving abilities, but struggle with complex tasks due to…
Agentic DraCor and the Art of Docstring Engineering: Evaluating MCP-empowered LLM Usage of the DraCor API
arXiv:2508.13774v1 Announce Type: cross Abstract: This paper reports on the implementation and evaluation of a Model Context Protocol (MCP) server…
PENGUIN: Enhancing Transformer with Periodic-Nested Group Attention for Long-term Time Series Forecasting
arXiv:2508.13773v1 Announce Type: cross Abstract: Long-term time series forecasting (LTSF) is a fundamental task with wide-ranging applications. Although Transformer-based models…
Note on Selection Bias in Observational Estimates of Algorithmic Progress
arXiv:2508.11033v2 Announce Type: replace-cross Abstract: Ho et. al (2024) attempts to estimate the degree of algorithmic progress from language models.…
SSPO: Self-traced Step-wise Preference Optimization for Process Supervision and Reasoning Compression
arXiv:2508.12604v1 Announce Type: cross Abstract: Test-time scaling has proven effective in further enhancing the performance of pretrained Large Language Models…
OpenMoCap: Rethinking Optical Motion Capture under Real-world Occlusion
arXiv:2508.12610v1 Announce Type: cross Abstract: Optical motion capture is a foundational technology driving advancements in cutting-edge fields such as virtual…
Visual Perception Engine: Fast and Flexible Multi-Head Inference for Robotic Vision Tasks
arXiv:2508.11584v2 Announce Type: replace-cross Abstract: Deploying multiple machine learning models on resource-constrained robotic platforms for different perception tasks often results…
Finite Automata Extraction: Low-data World Model Learning as Programs from Gameplay Video
arXiv:2508.11836v1 Announce Type: new Abstract: World models are defined as a compressed spatial and temporal learned representation of an environment.…
G-CUT3R: Guided 3D Reconstruction with Camera and Depth Prior Integration
arXiv:2508.11379v1 Announce Type: cross Abstract: We introduce G-CUT3R, a novel feed-forward approach for guided 3D scene reconstruction that enhances the…