Cross-Environment Neural Reranking for Sample-Efficient Action Selection in Text-Based Agents

• URL: http://arxiv.org/abs/2606.02204v1
• Date: Mon, 01 Jun 2026 13:00:58 GMT
• Title: Cross-Environment Neural Reranking for Sample-Efficient Action Selection in Text-Based Agents
• Abstract summary: Large language model agents achieve strong performance on text-based benchmarks but incur prohibitive inference costs.<n>We investigate whether a single lightweight model can perform action selection across multiple diverse environments.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language model agents achieve strong performance on text-based benchmarks but incur prohibitive inference costs, motivating the use of compact neural rerankers for action selection. We investigate whether a single lightweight model can perform action selection across multiple diverse environments, a capability that would eliminate per-environment model maintenance. Training DeBERTa-v3 (184M-434M parameters) jointly on ALFWorld, WebShop, and ScienceWorld with minority-class upsampling, we find that rebalanced two-environment joint training substantially improves over single-environment ALFWorld performance (net gain +0.412) while maintaining competitive WebShop performance (+0.214 vs. +0.249 single-environment). Three-environment training yields a mean combined net gain of +0.551 +/- 0.024 across 4 seeds, with per-environment results approaching specialized single-environment models while providing positive cross-domain transfer. Cross-environment adaptation is highly sample-efficient: fine-tuning on only 9.2% of target-domain data recovers 93% of full-data performance, and scaling model capacity yields limited benefits, indicating data diversity is the primary driver. Environment-aware LoRA adapter routing with PCGrad achieves a best-seed result of +0.611 (seed 42), with seeds 456 and 789 at +0.554 and +0.559, but exhibits high variance due to seed 123 collapsing to +0.263 (4-seed mean +0.497 +/- 0.158), representing a promising but currently unstable direction. Joint training with clean splits and data rebalancing is a key ingredient. We will release our three-environment benchmark of 51,580 training instances (41,740 raw unique states with minority-class upsampling) and all model checkpoints upon acceptance.

Related papers

• ER-KANs: Efficient and Robust Kolmogorov-Arnold Networks for Data-Scarce Scientific Machine Learning [0.0]
  We show that ChebyKAN's test MSE increases by a factor of 10.6x when training data is corrupted with sigma=0.1 noise.<n> ER-KAN combines three design choices targeting the noisy, data-scarce setting.
  arXiv  Detail & Related papers  (2026-08-14T16:15:37Z)
• ProCUA-SFT Technical Report [80.97543110323542]
  Training computer-use agents (CUAs) requires large-scale, diverse trajectory data collected in full desktop environments.<n>We present ProCUA-SFT, a dataset of 3.1M step-level SFT samples distilled from 93K synthetic trajectories.<n>Fine-tuning UI-TARS 7B on ProCUA-SFT for one epoch yields 45.0% on OSWorld -- an 18.7 percentage-point improvement over the base model.
  arXiv  Detail & Related papers  (2026-06-15T22:04:11Z)
• Spokes: Optimizing for Diverse Pretraining Data Selection [82.66872118512403]
  We introduce a probabilistic diversification framework based on the G-Vendi score, optimized via exponentiated gradient descent.<n>Our method produces subsets substantially more diverse than those obtained via random sampling.<n>We evaluate our approach on FineWeb and DCLM, where it consistently outperforms existing methods.
  arXiv  Detail & Related papers  (2026-06-13T09:28:46Z)
• Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization [93.59040719904154]
  This paper introduces RACES (textbfRecursive textbfAutomated textbfComposition for textbfEnvironment textbfScaling), a framework that conceptualizes verifiable environments as composable building blocks.<n> RACES is implemented with 300 individual environments and defines a set of composition operators that induce diverse reasoning patterns.
  arXiv  Detail & Related papers  (2026-06-10T17:39:41Z)
• TRACE: Capability-Targeted Agentic Training [11.088296897059157]
  We introduce TRACE, an end-to-end system for environment-specific agent self-improvement.<n>TRACE contrasts successful and failed trajectories to identify lacking capabilities.<n>It synthesizes a targeted training environment for each that rewards whether the capability was exercised.<n>It trains a LoRA adapter via RL on each synthetic environment, routing to the relevant adapter at inference.
  arXiv  Detail & Related papers  (2026-04-07T02:22:44Z)
• Fusion Complexity Inversion: Why Simpler Cross View Modules Outperform SSMs and Cross View Attention Transformers for Pasture Biomass Regression [0.0]
  adaptation of vision foundation models to agricultural regression is systematically evaluated on the CSIRO Pasture Biomass benchmark.<n>On scarce agricultural data, a two layer gated depthwise convolution (R2 = 0.903) outperforms cross view attention transformers.<n>Backbone pretraining scale is found to monotonically dominate all architectural choices.
  arXiv  Detail & Related papers  (2026-03-08T21:41:01Z)
• Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning [62.499592503950026]
  Large language model (LLM) have empowered autonomous agents to perform complex tasks that require multi-turn interactions with tools and environments.<n>We propose Agent World Model (AWM), a fully synthetic environment generation pipeline.<n>We scale to 1,000 environments covering everyday scenarios, in which agents can interact with rich toolsets.
  arXiv  Detail & Related papers  (2026-02-10T18:55:41Z)
• Extract-0: A Specialized Language Model for Document Information Extraction [0.0]
  This paper presents Extract-0, a 7-billion parameter language model specifically optimized for document information extraction.<n>Extract-0 achieves a mean reward of 0.573 on a benchmark of 1,000 diverse document extraction tasks, outperforming GPT-4.1 (0.457), o3 (0.464), and GPT-4.1-2025 (0.459).
  arXiv  Detail & Related papers  (2025-09-26T20:34:43Z)
• Bridging Synthetic and Real-World Domains: A Human-in-the-Loop Weakly-Supervised Framework for Industrial Toxic Emission Segmentation [2.6954348706500766]
  Industrial smoke segmentation is critical for air-quality monitoring and environmental protection but is often hampered by the high cost and scarcity of pixel-level annotations in real-world settings.<n>We introduce CEDANet, a class-aware domain adaptation framework that integrates weak, citizen-provided video-level labels with adversarial feature alignment.<n>CEDANet achieves an F1-score of 0.414 and a smoke-class IoU of 0.261 with citizen feedback, vastly outperforming the baseline model.
  arXiv  Detail & Related papers  (2025-07-29T16:53:00Z)
• FlowTS: Time Series Generation via Rectified Flow [67.41208519939626]
  FlowTS is an ODE-based model that leverages rectified flow with straight-line transport in probability space.<n>For unconditional setting, FlowTS achieves state-of-the-art performance, with context FID scores of 0.019 and 0.011 on Stock and ETTh datasets.<n>For conditional setting, we have achieved superior performance in solar forecasting.
  arXiv  Detail & Related papers  (2024-11-12T03:03:23Z)
• Towards Instance-adaptive Inference for Federated Learning [80.38701896056828]
  Federated learning (FL) is a distributed learning paradigm that enables multiple clients to learn a powerful global model by aggregating local training. In this paper, we present a novel FL algorithm, i.e., FedIns, to handle intra-client data heterogeneity by enabling instance-adaptive inference in the FL framework. Our experiments show that our FedIns outperforms state-of-the-art FL algorithms, e.g., a 6.64% improvement against the top-performing method with less than 15% communication cost on Tiny-ImageNet.
  arXiv  Detail & Related papers  (2023-08-11T09:58:47Z)
• Rethinking InfoNCE: How Many Negative Samples Do You Need? [54.146208195806636]
  We study how many negative samples are optimal for InfoNCE in different scenarios via a semi-quantitative theoretical framework. We estimate the optimal negative sampling ratio using the $K$ value that maximizes the training effectiveness function.
  arXiv  Detail & Related papers  (2021-05-27T08:38:29Z)

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.