2026· Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)· pp. 563-584· 0 citations· 27 references
TL;DR
This work proposes a closed-loop Multi-Agent Simulation Framework to synthe-size diverse, faithful, and policy-aligned shopping trajectories, and presents synthetic data that enables a small model to significantly outperform same-size baselines and surpass a large-model baseline.
Abstract
Modern e-commerce assistants must go beyond simple product search to support inspiration, comparison, and tool-grounded fact-checking across non-linear shopping journeys. However, distilling these complex behaviors into efficient, deployable models is bottle-necked by a lack of post-training data: trajectories must cover diverse agentic workflows with high fidelity, yet the desired outputs are open-ended without a single ground truth. We propose a closed-loop Multi-Agent Simulation Framework to synthe-size diverse, faithful, and policy-aligned shopping trajectories. The system orchestrates a journey-aware, stateful user simulator to drive exploration, a shopping agent that manages both tools and UI elements, and a critic agent that provides rubric-driven feedback to iteratively refine the data. On a domain-specific benchmark, this synthetic data enables a small model to significantly outperform same-size baselines and surpass a large-model baseline, making it viable to deploy the smaller back-bone at 8 × higher inference throughput than the large baseline.
See, a two-stage data synthesis framework consisting of an efficient exploration stage that builds an explicit UI transition graph over screens and elements, and a graph-based synthesis stage that composes diverse multi-step trajectories via planning and controlled sampling, yields reproducible and explainable data generation.
Zhuohang Fan, Beichen Zhang, Yuanfa Li et al.· 0 citations
Agentic search enables large language models to solve knowledge-intensive tasks by interleaving multi-step reasoning with retrieval, yet optimizing this with outcome-based reinforcement learning (RL) provides only sparse supervision. Knowledge distillation can supply denser guidance, and advanced proprietary models with their strong reasoning capabilities are promising teachers. While distilling from proprietary models can densify this supervisory signal, conventional logit-matching is precluded by hidden logits and mismatched tokenizers, whereas raw natural language trajectory imitation transfers superficial stylistic artifacts rather than core reasoning competence. To address the heterogeneous distillation problem and bridge the distribution gap, we propose Multi-Agent Protocol Distillation (MAPD), a joint distillation and RL framework uses a structured, style-normalized protocol as an intermediate representation. An offline multi-agent system (MAS) decomposes each query, retrieves supporting evidence, repairs failed searches, and converts the resulting exploration trace into a JSON protocol containing the task type, reasoning plan, and extractive grounding facts. During training, the protocol is provided only to a privileged branch of the student policy, whose token distributions furnish a dense distillation signal alongside the sparse RL objective. Extensive evaluations across seven QA benchmarks demonstrate that MAPD consistently outperforms competitive distillation and RL, achieving average success rates of 39.4\% on Qwen3-1.7B and 44.4\% on Qwen3-4B. Crucially, the framework generalizes robustly across diverse proprietary teachers while effectively mitigating the student policy from style drift and verbosity degeneration.
Junlin Liu, Jiangwang Chen, Zixin Song et al.· 4 citations
This research bridges theoretical foundations of reinforcement learning and graph-based memory with autonomous agent workflows, and offers a practical, scalable reference framework for developing artificial intelligence technologies in complex, multi-step autonomous systems.
Autonomous embodied agents must sustain a long decision-making loop that involves perceiving, acting, verifying, and self-correcting over many steps. Current systems sustain this loop through task-specific workflows or embodied policies. However, these fixed workflows and policies offer limited flexibility across environments and often lack effective recovery strategies when execution goes wrong. We find that a general-purpose agent can instead sustain the loop on its own. We term this organization agentic embodied control: the reasoning model directly steers every action, keeping reasoning and control aligned. Using zero-shot navigation as a controlled testbed, we equip three coding-agent harnesses with only a monocular RGB camera and discrete actions. At default effort, replicated opus-5 runs average $70.7\pm3.5$% success, while fable-5 reaches 78% at maximum effort. When a trained waypoint tool is offered alongside primitives, the hybrid fable-5 agent reaches $76.7\pm0.6$% at default effort, using half the environment steps and under a quarter of the wall time. Across the ablations, model choice dominates performance variation. Observed harness differences are modest, and forced waypoints help weaker models but can hinder stronger ones. Although longer horizons, latency, and context growth remain barriers to sustained autonomy, these results show that a general-purpose model can already achieve competitive embodied control without a navigation policy.
Jian Zhou, Xunyi Zhao, G. Zhou et al.· 0 citations
ODYSSE is presented, a Reinforced Fine-Tuning (RFT) framework for personalized agentic reasoning designed to address long action horizons and strong cross-step dependencies in personalized agentic reasoning, and an episodic batch sampler that groups actions from the same episode into unified training batches, facilitating coherent optimization under ESPO.
Jiaqi Zhang, Tong Chen, Junliang Yu et al.· 0 citations
Tool-Integrated Reasoning (TIR) enables LLMs to solve complex tasks through iterative tool interactions. However, existing reinforcement learning methods often rely on trajectory-level supervision, limiting fine-grained credit assignment in long-horizon TIR scenarios. On-policy self-distillation offers denser signals through teacher branches with privileged context, but existing approaches typically derive such context from ground-truth answers or retrieved skills, which may not reflect the states actually visited by the agent. Moreover, token-level supervision fails to capture the turn-level structure of tool interactions. To address this, we propose TurnSight, a turn-level hindsight self-distillation framework that derives supervision directly from execution-conditioned hindsight. It then constructs multiple hindsight views with different lookahead horizons and selects reliable supervision through cross-horizon directional agreement. Finally, the selected hindsight signal is normalized across sibling rollouts and used to adaptively modulate RL advantages while preserving their original optimization direction. Extensive experiments on three benchmarks demonstrate the effectiveness of TurnSight. Our codes are available at https://github.com/quchangle1/TurnSight.
Changle Qu, Sunhao Dai, Hengyi Cai et al.· 0 citations