Papers

• From Agent Failures to Text Policies: What Works and What Breaks (2026)
  Jaideep Ray et al.
  5.01
• PersonaTrail: Benchmarking Personalized Web Agents through Browsing Trails (2026)
  Seungbin Yang et al.
  4.39
• Interpretation-Oriented Cloud Removal via Observation-Anchored Residual Flow with Geo-Contextual Alignment (2026)
  Ziyao Wang et al.
  2.00
• TREK: Distill to Explore, Reinforce to Refine (2026)
  Yuanda Xu et al.
  2.00
• RoboTALES: Learning Reasoning-Guided Robot Policies via Task-Aligned Simulated Futures (2026)
  Hanan Gani et al.
  2.00
• Vision as Unified Multimodal Generation (2026)
  Xiaoyang Han et al.
  2.00
• Calibrate-Then-Act: Cost-Aware Exploration in LLM Agents (2026)
  Wenxuan Ding et al.
  1.94
• NGM: A Plug-and-Play Training-Free Memory Module for LLMs (2026)
  Yuwen Qu et al.
  1.94
• Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models (2026)
  Aswin RRV et al.
  1.89
• RePro: Training Language Models to Faithfully Recycle the Web for Pretraining (2025)
  Zichun Yu et al.
  1.50
• RExBench: Can coding agents autonomously implement AI research extensions? (2025)
  Nicholas Edwards et al.
  1.28
• Hybrid Code Networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning (2017)
  Jason D. Williams et al.
  —
• AnyTOD: A Programmable Task-Oriented Dialog System (2022)
  Jeffrey Zhao et al.
  —
• SGP-TOD: Building Task Bots Effortlessly via Schema-Guided LLM Prompting (2023)
  Xiaoying Zhang et al.
  —
• TANGO: Training-free Embodied AI Agents for Open-world Tasks (2024)
  Filippo Ziliotto et al.
  —
• WebArena: A Realistic Web Environment for Building Autonomous Agents (2023)
  Shuyan Zhou et al.
  —
• Lemur: Harmonizing Natural Language and Code for Language Agents (2023)
  Yiheng Xu et al.
  —
• MTU-Bench: A Multi-granularity Tool-Use Benchmark for Large Language Models (2024)
  Pei Wang et al.
  —
• WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning (2024)
  Zehan Qi et al.
  —