A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Yan Deng, Fei Xu
DreamFly introduces causal memory and receding-horizon diffusion planning to improve aerial vision-language navigation, achieving state-of-the-art results on the OpenFly benchmark.
Guancheng Zeng, Xueyi Chen, Jiawang Hu, et al.
Routine is a structured multi-step agent planning framework that significantly improves tool-calling accuracy in enterprise scenarios, boosting GPT-4o from 41.1% to 96.3% and enabling smaller models to approach that performance via fine-tuning.
Kuan Xing, Ye Wang, Changyi Gan, et al.
Atelier is a shortcut-aware control-state planning framework that improves artist-grounded text-to-image generation by explicitly modeling artistic intent and avoiding canonical style shortcuts.
Daniel Armstrong, Xuan-Vu Nguyen, Octavian Susanu, et al.
SynthEx, an agentic LLM framework, plans retrosynthetic routes to complex natural products that expert chemists judge comparable to human-designed syntheses.
Unknown
This paper demonstrates that world models can leverage exploratory data to plan or improve navigation policies, shifting focus from policy learning to model-based planning.
Unknown
HDFlow introduces a hierarchical diffusion-flow planning framework for long-horizon robotic assembly, leveraging a structured latent space to improve task decomposition and execution.
Unknown
This paper introduces a scalable safe long-horizon planning method that uses copulas to model temporal correlations in uncertainty, providing probabilistic guarantees via conformal prediction.
Unknown
This paper presents a language-guided long-horizon manipulation framework for multistep cloth folding, combining LLM-based planning with visual perception.
Unknown
This paper introduces structured preference optimization to generate and plan long-horizon tasks in vision-language models, improving task decomposition and execution.
Unknown
This paper introduces Lamma-p, a framework that integrates LLMs with a PDDL planner to enable generalizable multi-agent long-horizon task allocation and planning.
Unknown
This paper describes research on AI agents embodied in the world, focusing on modeling and planning to enable these agents to act effectively.
Unknown
ELHPlan introduces Action Chains, sequences of actions bound to sub-goal intentions, for efficient long-horizon multi-agent task planning.