…Generated by Qwen/Qwen2.5-Coder-32B-Instruct LLM agents are increasingly deployed as systems built around editable external harnesses, including prompts, skills, memories and tools, that shape task execution without changing…
…Agent skills are increasingly deployed as modular, reusable capability units in AI agent systems. Medical research agent skills require safeguards beyond general-purpose evaluation, including scientific integrity, methodological validity, reproducibility, and boundary…
…Programmable Runtime Skills for Efficient and Accurate LLM Agents (2026) RubricRefine: Improving Tool-Use Agent Reliability with Training-Free Pre-Execution Refinement (2026) RealClawBench: Live OpenClaw Benchmarks from Real Developer-Agent Sessions…
…to 24% — even with access to a tool it can use to ask for help. HiL-Bench is a benchmark that measures how well agents can identify when they're missing important…
…A Joint Tool-Prompt Reflective Optimization Framework for Language Agents (2026) Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programs (2026) CROP: Token-Efficient Reasoning in Large…
…tools, such as interactive maps for geography and audio for music. We evaluate Data2Story on 18 articles, each paired with the originally published expert piece, along four axes: (a) human-agent angle…
…Improving Review Substantiveness with Rubric-Guided, Tool-Integrated Agents (2026) EGTR-Review: Efficient Evidence-Grounded Scientific Peer Review Generation via Multi-Agent Teacher Distillation (2026) Can AI Review Improve Paper Drafting? An…
…A Diagnostic Framework for Planning Capabilities in LLM Agents (2026) ADAPT: Benchmarking Commonsense Planning under Unspecified Affordance Constraints (2026) EgoBench: An Interactive Egocentric Multimodal Benchmark for Tool-Using Agents (2026) PlanningBench: Generating…
…Benchmarking Agentic Procedural 3D Modeling Via Code Published on May 31 Submitted by taesiri on Jun 2 Deepmind Authors: Yipeng Gao , , , , , , , Abstract Vision-language models are evaluated for procedural 3D modeling tasks…
…Rule-based tools are blind to rendered visuals, operating only on source code and log files. Text-only LLMs perform open-loop text editing, unable to predict or verify the two-dimensional…