PANDO: Efficient Multimodal AI Agents via Online Skill Distillation

• URL: http://arxiv.org/abs/2605.24785v2
• Date: Tue, 26 May 2026 04:25:32 GMT
• Title: PANDO: Efficient Multimodal AI Agents via Online Skill Distillation
• Abstract summary: We introduce PANDO, a single-rollout online skill-distillation framework.<n>On the full set of 910 VisualWebArena tasks, PANDO achieves a 58.3% success rate.<n>A 300-task ablation shows that rules and routines provide most of the success gains.
• Score: 8.784808015938543
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in multimodal web agents often rely on increased inference-time computation, including rollout search, verifier passes, offline skill discovery, and specialist model stacks. This raises a central question: can a web agent become more efficient as it accumulates experience, rather than more expensive? We first analyze trajectories from VisualWebArena and identify three recurring sources of inefficiency: repeat-action loops, hidden discovery costs, and low prompt-cache reuse. We then introduce PANDO, a single-rollout online skill-distillation framework that maintains a structured Skill Library and combines progress reflection, confidence-based skill demotion, hierarchical routing, visual compression, and cache-aware prompting. On the full set of 910 VisualWebArena tasks, PANDO achieves a 58.3% success rate, outperforming SGV (54.0%) and our WALT reproduction (45.2%), while using 58% fewer tokens than SGV and 61% fewer tokens than WALT, without any pre-evaluation discovery budget. A 300-task ablation further shows that rules and routines provide most of the success gains, while routing, compression, and cache-aware prompting convert the larger skill library into lower marginal token cost. Finally, we introduce three trajectory-level efficiency metrics -- Action Repetition Rate, Step Overhead Ratio, and Prompt Cache Utilization -- to make efficiency visible beyond terminal success.

Related papers

• TRIAGE: Three-level Routing and Intelligent Agent Guidance for Efficient Execution [0.0]
  TRIAGE is a three-level routing framework that reduces token consumption by reusing historical execution trajectories.<n>Its core innovation is Tjectory-as-a-Skill, which abstracts historical execution trajectories into reusable skills.<n>In large-scale experiments on 1,007 security monitoring queries, TRIAGE achieves 62.3% token savings.
  arXiv  Detail & Related papers  (2026-09-01T15:40:09Z)
• PolicyTrim: Boosting Intrinsic Policy Efficiency of Vision-Language-Action Models [50.232333672172395]
  Vision-Language-Action (VLA) models provide a unified paradigm for robotic manipulation, yet their real-world deployment is often bottlenecked by execution efficiency.<n>We propose textbfPolicyTrim, a reinforcement learning-based post-training framework that extends the reliable action chunk length and reduces redundant physical steps.<n>Our framework delivers up to a 5.83$times$ end-to-end deployment speedup without compromising task success rates.
  arXiv  Detail & Related papers  (2026-06-21T14:54:07Z)
• CacheRL:Multi-Turn Tool-Calling Agents via Cached Rollouts and Hybrid Reward [17.939095034578035]
  CacheRL is a system for training small agent foundation models.<n>It achieves 92 percent process accuracy on multi-step tool-calling tasks.<n>It approaches GPT-5's 94 percent while requiring 100 times less compute.
  arXiv  Detail & Related papers  (2026-06-12T07:01:50Z)
• HyperEyes: Dual-Grained Efficiency-Aware Reinforcement Learning for Parallel Multimodal Search Agents [6.536862833942476]
  We present HyperEyes, a parallel multimodal search agent that fuses visual grounding and retrieval into a single atomic action.<n>Across six benchmarks, HyperEyes-30B surpasses the strongest comparable open-source agent by 9.9% in accuracy with 5.3x fewer tool-call rounds on average.
  arXiv  Detail & Related papers  (2026-05-08T03:16:08Z)
• M$^2$: Dual-Memory Augmentation for Long-Horizon Web Agents via Trajectory Summarization and Insight Retrieval [64.06936170117943]
  M$2$ is a training-free, memory-augmented framework designed to optimize context efficiency and decision-making.<n>Our approach incorporates a dual-tier memory mechanism that synergizes Dynamic Trajectory Summarization (Internal Memory) to compress verbose interaction history into concise state updates, and Insight Retrieval Augmentation (External Memory) to guide the agent with actionable guidelines retrieved from an offline insight bank.
  arXiv  Detail & Related papers  (2026-02-28T06:59:51Z)
• DEPO: Dual-Efficiency Preference Optimization for LLM Agents [75.6723341304463]
  We propose DEPO, a dual-efficiency preference optimization method that jointly rewards succinct responses and fewer action steps.<n>Experiments on WebShop and BabyAI show that DEPO cuts token usage by up to 60.9% and steps by up to 26.9%, while achieving up to a 29.3% improvement in performance.
  arXiv  Detail & Related papers  (2025-11-19T12:38:43Z)
• Branch-and-Browse: Efficient and Controllable Web Exploration with Tree-Structured Reasoning and Action Memory [69.49061918994882]
  Branch-and-Browse is a fine-grained web agent framework that unifies structured reasoning-acting, contextual memory, and efficient execution.<n>On the WebArena benchmark, Branch-and-Browse achieves a task success rate of 35.8% and reduces execution time by up to 40.4% relative to state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-10-18T00:45:37Z)
• Multi-Agent Reinforcement Learning for Sample-Efficient Deep Neural Network Mapping [54.65536245955678]
  We present a decentralized multi-agent reinforcement learning (MARL) framework designed to overcome the challenge of sample inefficiency.<n>We introduce an agent clustering algorithm that assigns similar mapping parameters to the same agents based on correlation analysis.<n> Experimental results show our MARL approach improves sample efficiency by 30-300x over standard single-agent RL.
  arXiv  Detail & Related papers  (2025-07-22T05:51:07Z)
• How to Train Your LLM Web Agent: A Statistical Diagnosis [96.86317871461834]
  We present the first statistically grounded study on compute allocation for LLM web-agent post-training.<n>Our approach uses a two-stage pipeline, training a Llama 3.1 8B student to imitate a Llama 3.3 70B teacher via supervised fine-tuning (SFT) and on-policy reinforcement learning.<n>Our results show that combining SFT with on-policy RL consistently outperforms either approach alone on both WorkArena and MiniWob++.
  arXiv  Detail & Related papers  (2025-07-05T17:12:33Z)
• Revisiting Token Pruning for Object Detection and Instance Segmentation [25.3324628669201]
  We investigate token pruning to accelerate inference for object and instance segmentation. We show a reduction in performance decline from 1.5 mAP to 0.3 mAP in both boxes and masks, compared to existing token pruning methods.
  arXiv  Detail & Related papers  (2023-06-12T11:55:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.