AI Agent Papers2026
Yansong Ning, Jingwen Ye +6Affiliation: AI Thrust, HKUST(GZ), TEG AIPD, Tencentyning092connect.hkust-gz.edu.cn, [email protected]{jingwenye,wadewdzhang}@tencent.com
Evaluates and trains multimodal agents to construct interactive 3D open worlds from user queries — provides a large benchmark of assets, seed worlds, and reverse-synthesized queries plus a sandbox RL gym for tool-driven editing and rubric-based verification. Reports that frontier MLLMs perform under 60% and that RL fine-tuning improves precise 3D editing.
AI Agent Papers2026
Trains compact conversational agents to adapt at runtime to changing 'Harness' configurations (Skills, Hooks, prompts, tools) using Harness-Aware Training (HAT): Harness-State Augmentation, on-policy distillation, and RL to preserve generality while meeting low-latency deployment constraints.
AI Agent Papers2026
Trains a foundation GUI agent using a closed-loop, environment-grounded data stack plus in-context multimodal demonstrations to automate long-horizon desktop workflows. Combines scalable task generation/verification, subtask-level demo guidance, and a 100-task OSWorkerBench benchmark to improve strict success and task progress.
Presents a unified black-box reinforcement learning framework to train and optimize agents running inside complex execution harnesses. Uses sandbox-parallel rollouts, a serving proxy that captures model calls and reconstructs multi-turn trajectories as prefix trees, and adapted GRPO/PPO optimizers to achieve stable, scalable RL across heterogeneous harnesses.
AI Agent Papers2026
Enables closed-loop execution for embodied agents by evolving code-based runtime critics and recovery skills online while keeping the base policy frozen. Combines three timescale loops with Z-Infra rollout infrastructure; reports 90.8% on LIBERO-Pro, 93.6% on RoboCasa and an 11.1× inference speedup.
AI Agent Papers2026
Evaluates whether AI systems can independently carry out project-level scientific research by progressively removing human methodological guidance across 60 tasks in 11 domains. Built with expert review, sandbox execution, and multi-agent–model scoring to measure innovation and autonomous experimental execution.
Embodied AI2026
Turns embodied navigation into 2D visual prompting where a vision-language model selects image pixels that are projected to 3D actions; adds selective chain-of-thought, compressed anchor-trajectory memory, and a two-level alignment objective to improve sample and runtime efficiency.
Fine-tunes long-horizon LLM agents with evolution strategies so full-model updates run at inference-level GPU memory. Emphasizes trajectory-level credit via black-box rewards, online prompt–parameter co-evolution, and a cosine decay for perturbation scale to balance exploration and adaptation; suited for limited-GPU settings.
AI Agent Papers2026
Turns natural-language PLC requirements into verified, runnable IEC 61131-3 Structured Text by driving a closed loop of generation, compilation, deployment, and behavioral verification on a live OpenPLC runtime. The verification-gated harness forces inputs, traces execution, repairs failures, and renders ladder diagrams plus process simulation to raise dynamic runtime pass rates.
AI Agent Papers2026
Proposes FACET, a framework that synthesizes verifiable terminal tasks by reconstructing scenario intent and grounding instruction, solution, and verifier in a shared executable container state. Key features include environment-first generation, execution-based validation, and targeted repair to preserve source intent and cross-artifact consistency.
AI Agent Papers2026
Wraps static, hand-built environments with a programmable plug-in harness that reshapes environment behavior without changing underlying logic. EnvRigger automates diagnosis and synthesis of harness components from agent failure trajectories, validating edits via fresh rollouts to improve agent success and efficiency.
AI Agent Papers2026
Evaluates whether coding agents can modify real scientific software while preserving domain-specific scientific contracts. Contains 119 repository-level tasks across 98 GitHub projects and 20 scientific domains, measures reproducible edits in pinned Docker images, and analyzes recurring failure modes.