HFEPX Hub
Updated from current HFEPX corpus (Feb 27, 2026). 10 papers are grouped in this hub page. Common evaluation modes: Simulation Env, Automatic Metrics. Most common rater population: Domain Experts. Common annotation unit: Trajectory. Frequently cited benchmark: APPS. Common metric signal: cost. Newest paper in this set is from Feb 24, 2026.
Updated from current HFEPX corpus (Feb 27, 2026). This page covers 10 papers centered on Web Browsing + Simulation Env Papers. Common evaluation modes include Simulation Env, Automatic Metrics, with benchmark emphasis on APPS, BrowseComp. Use the anchored takeaways below to compare protocol choices and identify papers with stronger evidence depth.
Evaluation emphasis: Simulation Env and Automatic Metrics appear frequently in this slice.
Evidence: Efficient Hierarchical Any-Angle Path Planning on Multi-Resolution 3D Grids , SoK: Agentic Skills -- Beyond Tool Use in LLM Agents , Contextual Safety Reasoning and Grounding for Open-World Robots
Benchmark concentration: APPS, BrowseComp helps control cross-paper variance.
Evidence: SoK: Agentic Skills -- Beyond Tool Use in LLM Agents , Contextual Safety Reasoning and Grounding for Open-World Robots , MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks
Metric concentration: cost, accuracy is repeatedly reported in this group.
Evidence: Contextual Safety Reasoning and Grounding for Open-World Robots , MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks , BrowseComp-$V^3$: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents
Stratify by benchmark (APPS vs BrowseComp) before comparing methods.
Evidence: MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks , BrowseComp-$V^3$: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents , UI-Venus-1.5 Technical Report
Track metric sensitivity by reporting both cost and accuracy.
Evidence: BrowseComp-$V^3$: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents , UI-Venus-1.5 Technical Report , Aerial Vision-Language Navigation with a Unified Framework for Spatial, Temporal and Embodied Reasoning
Papers with explicit human feedback is visible in approximately 30% of papers in this set.
Evidence: UI-Venus-1.5 Technical Report , Aerial Vision-Language Navigation with a Unified Framework for Spatial, Temporal and Embodied Reasoning , MoMaGen: Generating Demonstrations under Soft and Hard Constraints for Multi-Step Bimanual Mobile Manipulation
Papers with explicit human feedback
Coverage is usable but incomplete (30% vs 45% target).
Papers reporting quality controls
Coverage is a replication risk (0% vs 30% target).
Papers naming benchmarks/datasets
Coverage is strong (40% vs 35% target).
Papers naming evaluation metrics
Coverage is strong (50% vs 35% target).
Papers with known rater population
Coverage is a replication risk (20% vs 35% target).
Papers with known annotation unit
Coverage is usable but incomplete (30% vs 35% target).
1. Efficient Hierarchical Any-Angle Path Planning on Multi-Resolution 3D Grids
Start with this anchor paper for scope and protocol framing. Covers Simulation Env.
2. SoK: Agentic Skills -- Beyond Tool Use in LLM Agents
Covers Simulation Env.
3. Contextual Safety Reasoning and Grounding for Open-World Robots
Covers Simulation Env.
4. MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks
Covers Simulation Env. Includes human-feedback signal: Pairwise Preference.
5. BrowseComp-$V^3$: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents
Covers Automatic Metrics, Simulation Env.
6. UI-Venus-1.5 Technical Report
Covers Simulation Env.
Covers Simulation Env.
Covers Simulation Env. Includes human-feedback signal: Demonstrations.
simulation_env vs automatic_metrics
both=1, left_only=9, right_only=0
1 papers use both Simulation Env and Automatic Metrics.
Victor Reijgwart, Cesar Cadena, Roland Siegwart, Lionel Ott · Feb 24, 2026 · Citations: 0
Hierarchical, multi-resolution volumetric mapping approaches are widely used to represent large and complex environments as they can efficiently capture their occupancy and connectivity information.
Yanna Jiang, Delong Li, Haiyu Deng, Baihe Ma, Xu Wang · Feb 24, 2026 · Citations: 0
Agentic systems increasingly rely on reusable procedural capabilities, \textit{a.k.a., agentic skills}, to execute long-horizon workflows reliably.
Zachary Ravichandran, David Snyder, Alexander Robey, Hamed Hassani, Vijay Kumar · Feb 23, 2026 · Citations: 0
Traditional safety approaches enforce fixed constraints in user-specified contexts, limiting their ability to handle the open-ended contextual variability of real-world deployment.
Zexue He, Yu Wang, Churan Zhi, Yuanzhe Hu, Tzu-Ping Chen · Feb 18, 2026 · Citations: 0
Existing evaluations of agents with memory typically assess memorization and action in isolation.
Huanyao Zhang, Jiepeng Zhou, Bo Li, Bowen Zhou, Yanzhe Shan · Feb 13, 2026 · Citations: 0
Multimodal large language models (MLLMs), equipped with increasingly advanced planning and tool-use capabilities, are evolving into autonomous agents capable of performing multimodal web browsing and deep search in open-world environments.
Venus Team, Changlong Gao, Zhangxuan Gu, Yulin Liu, Xinyu Qiu · Feb 9, 2026 · Citations: 0
GUI agents have emerged as a powerful paradigm for automating interactions in digital environments, yet achieving both broad generality and consistently strong task performance remains challenging.
Huilin Xu, Zhuoyang Liu, Yixiang Luomei, Feng Xu · Dec 9, 2025 · Citations: 0
Extensive experiments on the AerialVLN and OpenFly benchmark validate the effectiveness of our method.
Chengshu Li, Mengdi Xu, Arpit Bahety, Hang Yin, Yunfan Jiang · Oct 21, 2025 · Citations: 0
Imitation learning from large-scale, diverse human demonstrations has been shown to be effective for training robots, but collecting such data is costly and time-consuming.
Amin Seffo, Aladin Djuhera, Masataro Asai, Holger Boche · Jun 4, 2025 · Citations: 0
Recent advancements in large language models (LLMs) have spurred interest in robotic navigation that incorporates complex spatial, mathematical, and conditional constraints from natural language into the planning problem.
Zeyi Liao, Jaylen Jones, Linxi Jiang, Yuting Ning, Eric Fosler-Lussier · May 28, 2025 · Citations: 0
Computer-use agents (CUAs) promise to automate complex tasks across operating systems (OS) and the web, but remain vulnerable to indirect prompt injection.
