Push Smarter, Not Harder: Hierarchical RL-Diffusion Policy for Efficient Nonprehensile Manipulation

The recent advancements in generative AI for software engineering have led to the emergence of agentic coding, where AI systems can plan and execute code changes. However, many enterprise AI coding pilots are underperforming, primarily due to inadequate context surrounding the code, rather than flaws in the AI models themselves.

GitHub has announced updates to its AI app-generation tool, Spark, which is currently in public preview. The latest enhancements include improvements in enterprise capabilities, billing features, and user interface upgrades, aimed at streamlining the app-building process for developers.

A new study introduces a data-efficient fine-tuning strategy for large-scale text-to-video diffusion models, enabling the addition of generative controls over physical camera parameters using sparse, low-quality synthetic data. This approach demonstrates that models fine-tuned on simpler data can outperform those trained on high-fidelity datasets.

SplatCo has been introduced as a novel structure-view collaborative Gaussian splatting framework designed for high-fidelity rendering of complex outdoor scenes. This framework integrates a cross-structure collaboration module, a cross-view pruning mechanism, and a structure view co-learning module to enhance detail preservation and rendering efficiency in large-scale unbounded scenes.

A recent study explores the automated recognition of instructional activities and discourse from multimodal classroom data, utilizing AI-driven analysis of 164 hours of video and 68 lesson transcripts. This research aims to replace manual annotation methods, which are resource-intensive and difficult to scale, with more efficient AI techniques for actionable feedback to educators.

The introduction of the D³-Predictor presents a significant advancement in dense prediction by addressing the limitations of existing diffusion models, which are hindered by stochastic noise that disrupts fine-grained spatial cues and geometric structure mappings. This new framework reformulates a pretrained diffusion model to eliminate stochasticity, allowing for a more deterministic mapping from images to geometry.

A new systematic pipeline has been established for classifying biologically relevant light contexts from wearable data, utilizing ActLumus recordings from 26 participants over a week. The pipeline includes steps such as domain selection, log-base-10 transformation, and L2 normalization, achieving high performance in distinguishing natural from artificial light.

A recent study has introduced differential smoothing as a method to mitigate the diversity collapse often observed in large language models (LLMs) during reinforcement learning fine-tuning. This method aims to enhance both the correctness and diversity of model outputs, addressing a critical issue where outputs lack variety and can lead to diminished performance across tasks.