Autonomous agents, tool use, and agentic systems
![[2510.16609] Prior Knowledge Makes It Possible: From Sublinear Graph Algorithms to LLM Test-Time Methods](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
Abstract page for arXiv paper 2510.16609: Prior Knowledge Makes It Possible: From Sublinear Graph Algorithms to LLM Test-Time Methods
Abstract page for arXiv paper 2604.02131: Intelligent Cloud Orchestration: A Hybrid Predictive and Heuristic Framework for Cost Optimization
Abstract page for arXiv paper 2604.01346: Safety, Security, and Cognitive Risks in World Models
The paper introduces FML-bench, a new benchmark for evaluating machine learning agents in scientific research, focusing on exploration di...
The paper presents SLM-MUX, a novel architecture for orchestrating small language models (SLMs) to improve reasoning accuracy, achieving ...
The paper presents EpidemIQs, a multi-agent framework utilizing large language models for efficient epidemic modeling, demonstrating impr...
This article explores the concept of Grounding IDs, which are latent identifiers that enhance multimodal binding in large vision-language...
This article explores the foundational bottlenecks in multimodal reasoning, highlighting how additional modalities can enhance or hinder ...
The EO-1 model is introduced as a unified foundation for general robot control, enhancing multimodal reasoning through a large dataset an...
The paper introduces ClearFairy, an AI assistant designed to enhance decision-making in creative workflows by structuring reasoning and i...
The paper presents CASCADE, a novel LLM-powered JavaScript deobfuscator developed by Google, which enhances code comprehension and analys...
The paper presents LeanPremise, a neural premise selection system that enhances LeanHammer, a tool for automated reasoning in proof assis...
The paper presents Knowledgeable-R1, a reinforcement-learning framework designed to enhance retrieval-augmented generation (RAG) by mitig...
The paper presents PROGRESS, a framework for prioritized concept learning in vision-language models, enabling efficient sample selection ...
The paper presents QCS-ADME, a novel quantum circuit search framework for predicting drug properties, addressing challenges in imbalanced...
The paper 'Renaissance' explores the pretraining of vision-language encoders, addressing best practices and introducing a flexible evalua...
This article presents a novel framework for managing transportation infrastructure using multi-agent deep reinforcement learning, address...
The paper discusses advancements in AI towards ultra-long-horizon autonomy, introducing ML-Master 2.0, which utilizes Hierarchical Cognit...
The paper introduces KeplerAgent, a physics-guided LLM framework designed for symbolic equation discovery, enhancing accuracy and robustn...
The paper presents a novel framework for part retrieval in 3D CAD assemblies using vision-language models, emphasizing training-free meth...
The paper presents a multi-agent framework to enhance contextual privacy in large language models (LLMs), demonstrating a significant red...
The paper presents InsightX Agent, an LMM-based framework that enhances X-ray non-destructive testing (NDT) by improving reliability, int...
The paper presents Med-REFL, a framework designed to enhance medical reasoning in AI by enabling self-correction through fine-grained ref...
Get the latest news, tools, and insights delivered to your inbox.
Daily or weekly digest • Unsubscribe anytime