AI 论文 · 2026-06
浏览 2026-06 发布的AI 论文内容，第 9 页，共 818 条。
- Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories
- OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents
- AutoMedBench: Towards Medical AutoResearch with Agentic AI Models
- WALL-WM: Carving World Action Modeling at the Event Joints
- Absorbing Complexity: An Interaction-Native Knowledge Harness for Financial LLM Agents
- LayerRoute: Input-Conditioned Adaptive Layer Skipping via LoRA Fine-Tuning for Agentic Language Models
- PlatonicNav: Unveiling Semantic Correspondence in Navigation with Platonic Topological Maps
- Quality-Guided Semi-Supervised Learning for Medical Image Segmentation
- Decentralized Instruction Tuning: Conflict-Aware Splitting and Weight Merging
- Two-Fidelity Best-Action Identification for Stochastic Minimax Tree
- Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning
- DOT-MoE: Differentiable Optimal Transport for MoEfication
- Multi-Agent Computer Use
- Joint Agent Memory and Exploration Learning via Novelty Signals
- Scalable Inference-Time Annealing with Surrogate Likelihood Estimators
- LLM Anonymization Against Agentic Re-Identification
- Semantic Motion Anchors: Bridging Motion and Meaning in Co-Speech Gestures
- Parametric Social Identity Injection and Diversification in Public Opinion Simulation