SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating

• URL: http://arxiv.org/abs/2606.07074v1
• Date: Fri, 05 Jun 2026 09:10:50 GMT
• Title: SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating
• Abstract summary: Deep research agents have demonstrated remarkable capabilities in complex information-seeking tasks, yet this power comes at a steep computational cost.<n>We propose SlimSearcher, a principled framework that pushes the frontier between accuracy and computational cost across Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL)<n>Experiments on long-horizon benchmarks, including GAIA, BrowseComp, and XBenchDeepSearch, demonstrate that SlimSearcher reduces average tool-call rounds by 17%-58% while maintaining or improving accuracy.
• Score: 26.487281765184083
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep research agents have demonstrated remarkable capabilities in complex information-seeking tasks, yet this power comes at a steep computational cost. Driven by accuracy-focused training paradigms, current models adopt brute-force strategies characterized by blind tool dependency and performative reasoning-generating long, redundant trajectories that are far from necessary for resolving these tasks, leading to wasteful tool calls and excessive token consumption. To overcome this efficiency trap, we propose SlimSearcher, a principled framework that pushes the Pareto frontier between accuracy and computational cost across both Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). In the SFT stage, SlimSearcher employs Pareto-efficient filtration to distill trajectories that are both successful and economical, guiding the model toward inherently efficiency-aware search behaviors. During RL, we introduce Adaptive Reward Gating, a dynamic reward-shaping mechanism that evaluates relative tool and token efficiency within a sampled cohort. By cascading these adaptive efficiency metrics with a strict correctness gate, our approach effectively avoids the brevity bias associated with absolute penalties and mitigates reward hacking. Extensive experiments on long-horizon benchmarks, including GAIA, BrowseComp, and XBenchDeepSearch, demonstrate that SlimSearcher reduces average tool-call rounds by 17%-58% while maintaining or improving accuracy.

Related papers

• ODAR: Principled Adaptive Routing for LLM Reasoning via Active Inference [60.958331943869126]
  ODAR-Expert is an adaptive routing framework that optimize the accuracy-efficiency trade-off via principled resource allocation.<n>We show strong and consistent gains, including 98.2% accuracy on MATH and 54.8% on Humanity's Last Exam.
  arXiv  Detail & Related papers  (2026-02-27T05:22:01Z)
• LightSearcher: Efficient DeepSearch via Experiential Memory [23.338677838845]
  We propose an efficient reinforcement learning framework that balances accuracy and efficiency in DeepSearch paradigms.<n>Experiments on four multi-hop QA benchmarks show that LightSearcher maintains accuracy comparable to SOTA baseline ReSearch.
  arXiv  Detail & Related papers  (2025-12-07T04:29:52Z)
• WebLeaper: Empowering Efficiency and Efficacy in WebAgent via Enabling Info-Rich Seeking [60.35109192765302]
  Information seeking is a core capability that enables autonomous reasoning and decision-making.<n>We propose WebLeaper, a framework for constructing high-coverage IS tasks and generating efficient solution trajectories.<n>Our method consistently achieves improvements in both effectiveness and efficiency over strong baselines.
  arXiv  Detail & Related papers  (2025-10-28T17:51:42Z)
• HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness [49.72591739116668]
  Reinforcement Learning (RL) has become a key driver for enhancing the long chain-of-thought (CoT) reasoning capabilities of Large Language Models (LLMs)<n>However, prevalent methods like GRPO often fail when task difficulty exceeds the model's capacity, leading to reward sparsity and inefficient training.<n>We propose HINT: Helping Ineffective rollouts Navigate Towards effectiveness, an adaptive hinting framework.
  arXiv  Detail & Related papers  (2025-10-10T13:42:03Z)
• EconProver: Towards More Economical Test-Time Scaling for Automated Theorem Proving [64.15371139980802]
  Large Language Models (LLMs) have recently advanced the field of Automated Theorem Proving (ATP)<n>We show that different test-time scaling strategies for ATP models introduce significant computational overhead for inference.<n>We propose two complementary methods that can be integrated into a unified EconRL pipeline for amplified benefits.
  arXiv  Detail & Related papers  (2025-09-16T03:00:13Z)
• How Far Are We from Optimal Reasoning Efficiency? [23.593914897406943]
  Large Reasoning Models (LRMs) demonstrate remarkable problem-solving capabilities through extended Chain-of-Thought (CoT) reasoning.<n>LRMs often produce excessively verbose and redundant reasoning traces.<n>Existing fine-tuning methods aim to improve reasoning efficiency, but assessing their efficiency gains remains challenging.
  arXiv  Detail & Related papers  (2025-06-08T12:18:50Z)
• RL-PINNs: Reinforcement Learning-Driven Adaptive Sampling for Efficient Training of PINNs [0.0]
  Physics-Informed Neural Networks (PINNs) have emerged as a powerful framework for solving partial differential equations (PDEs)<n>Their performance heavily relies on the strategy used to select training points.<n>We propose RL-PINNs, a reinforcement learning-driven adaptive sampling framework that enables efficient training with only a single round of sampling.
  arXiv  Detail & Related papers  (2025-04-17T13:50:55Z)
• BIPEFT: Budget-Guided Iterative Search for Parameter Efficient Fine-Tuning of Large Pretrained Language Models [63.52035708182815]
  We introduce a novel Budget-guided Iterative search strategy for automatic PEFT (BIPEFT) BIPEFT employs a new iterative search strategy to disentangle the binary module and rank dimension search spaces. Extensive experiments on public benchmarks demonstrate the superior performance of BIPEFT for downstream tasks with a low parameter budget.
  arXiv  Detail & Related papers  (2024-10-04T18:50:46Z)
• Augmenting Unsupervised Reinforcement Learning with Self-Reference [63.68018737038331]
  Humans possess the ability to draw on past experiences explicitly when learning new tasks. We propose the Self-Reference (SR) approach, an add-on module explicitly designed to leverage historical information. Our approach achieves state-of-the-art results in terms of Interquartile Mean (IQM) performance and Optimality Gap reduction on the Unsupervised Reinforcement Learning Benchmark.
  arXiv  Detail & Related papers  (2023-11-16T09:07:34Z)
• LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning [56.88751562302793]
  Low-rank adaption (LoRA) has emerged to fine-tune large language models (LLMs) LoRAPrune is a new framework that delivers an accurate structured pruned model in a highly memory-efficient manner. LoRAPrune achieves a reduction in perplexity by 4.81 on WikiText2 and 3.46 on PTB, while also decreasing memory usage by 52.6%.
  arXiv  Detail & Related papers  (2023-05-28T15:15:48Z)
• A Transferable and Automatic Tuning of Deep Reinforcement Learning for Cost Effective Phishing Detection [21.481974148873807]
  Many challenging real-world problems require the deployment of ensembles multiple complementary learning models. Deep Reinforcement Learning (DRL) offers a cost-effective alternative, where detectors are dynamically chosen based on the output of their predecessors.
  arXiv  Detail & Related papers  (2022-09-19T14:09:07Z)

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.