LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering
Loop Engineering is emerging as a practice for organizing development work around coding agents. Instead of writing each prompt by hand, practitioners design loops that monitor progress, assign work, run checks, and decide what the agent should do next. Even with a capable coding agent, a loop may trust a stale progress note, skip needed verification, spend its budget in the wrong direction, or stop before the task is safe to submit. Yet the final outcome of one end-to-end run cannot tell whether success or failure reflects the loop's guidance or the coding agent's ability to carry out the task. We introduce LoopArena, a benchmark for evaluating how well one model can guide a separate coding agent through a long-running task. The model under evaluation is the Controller: after each coding round, it receives a structured summary of the run and instructs a separate, fixed coding agent, the Worker, on what to do or verify next, or decides whether to stop. LoopArena evaluates this ability in three complementary settings that differ in execution scope and cost. Type I consists of 160 low-cost control questions; its execution cost is incurred once during benchmark construction, so each new Controller is evaluated through a single four-way choice with no code execution. Type II evaluates Controller guidance on 80 selected task slices, while Type III evaluates the corresponding 80 full tasks from their original states. On full tasks, the best observed Strict Success Rate is 36.25%, leaving substantial room for improvement in long-horizon loop control. Across Controllers, the paired reduction in estimated inference cost averages 60.6%, and Type II produces a positively correlated ordering under the main Core criterion (Spearman's rho = 0.7353). We release the benchmark data and evaluation code at AMAP-ML/LoopArena.
Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes
City-scale 3D grounding is commonly formulated as candidate ranking, which requires precomputed instance proposals and becomes inefficient in large outdoor scenes. We propose Urbex, an agentic vision-language framework that instead casts grounding as active spatial search. Urbex represents a city-scale 3D scene as an interactive multi-view environment, allowing a VLM agent to query landmarks, zoom into local regions, render oblique views, and commit a grounded bounding box. We optimize this tool-use policy with reinforcement learning on CityRefer, using smooth localization rewards and lightweight shaping rewards for efficient exploration. Experiments show that Urbex achieves the best Hit@1 and IoU on CityRefer, supports candidate-free grounding without ground-truth boxes, and is substantially faster than the strongest city-scale ranking baseline. Zero-shot results on STPLS3D-Refer further indicate improved cross-scene robustness.
Comments: NeurIPS 2026. 23 pages, 10 figures. Code: AMAP-ML/Urbex
Mentions
@Khazix0918 最近在用这个 AMAP-ML/LongHorizon-Harness 思路差不多
This AI Agent Runs Your Desktop For HOURS! 🤯 #Shorts LongHorizon-Harness is the game-changing framework for running AI agents across desktop apps and CLI for extended periods. Fresh-context execution, durable state, and native Claude Code integration. Check it out: AMAP-ML/LongHorizon-Harness #aiagent #claude #automation #github #techshorts #python #coding
@HuggingPapers Thanks for sharing our NeurIPS26 paper. We also release the code AMAP-ML/Urbex
LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering Loop Engineering is emerging as a practice for organizing development work around coding agents. Instead of writing each prompt by hand, practitioners design loops that monitor progress, assign work, run checks, and decide what the agent should do next. Even with a capable coding agent, a loop may trust a stale progress note, skip needed verification, spend its budget in the wrong direction, or stop before the task is safe to submit. Yet the final outcome of one end-to-end run cannot tell whether success or failure reflects the loop's guidance or the coding agent's ability to carry out the task. We introduce LoopArena, a benchmark for evaluating how well one model can guide a separate coding agent through a long-running task. The model under evaluation is the Controller: after each coding round, it receives a structured summary of the run and instructs a separate, fixed coding agent, the Worker, on what to do or verify next, or decides whether to stop. LoopArena evaluates this ability in three complementary settings that differ in execution scope and cost. Type I consists of 160 low-cost control questions; its execution cost is incurred once during benchmark construction, so each new Controller is evaluated through a single four-way choice with no code execution. Type II evaluates Controller guidance on 80 selected task slices, while Type III evaluates the corresponding 80 full tasks from their original states. On full tasks, the best observed Strict Success Rate is 36.25%, leaving substantial room for improvement in long-horizon loop control. Across Controllers, the paired reduction in estimated inference cost averages 60.6%, and Type II produces a positively correlated ordering under the main Core criterion (Spearman's rho = 0.7353). We release the benchmark data and evaluation code at AMAP-ML/LoopArena.
Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes City-scale 3D grounding is commonly formulated as candidate ranking, which requires precomputed instance proposals and becomes inefficient in large outdoor scenes. We propose Urbex, an agentic vision-language framework that instead casts grounding as active spatial search. Urbex represents a city-scale 3D scene as an interactive multi-view environment, allowing a VLM agent to query landmarks, zoom into local regions, render oblique views, and commit a grounded bounding box. We optimize this tool-use policy with reinforcement learning on CityRefer, using smooth localization rewards and lightweight shaping rewards for efficient exploration. Experiments show that Urbex achieves the best Hit@1 and IoU on CityRefer, supports candidate-free grounding without ground-truth boxes, and is substantially faster than the strongest city-scale ranking baseline. Zero-shot results on STPLS3D-Refer further indicate improved cross-scene robustness. Comments: NeurIPS 2026. 23 pages, 10 figures. Code: AMAP-ML/Urbex