arXiv AI

AHEAD: Adaptive Hindsight with Environment-Augmented Distillation for Agentic RL

AHEAD is a step‑aware framework that augments reinforcement learning for multi‑turn LLM agents by matching different supervision sources to different step types. The teacher receives environment feedback on all steps and LLM‑generated corrective hints only on error steps, providing finer‑grained guidance than uniform trajectory‑level rewards. Across ALFWorld, WebShop, and Search‑based QA, AHEAD improves task success by 13.3 points on ALFWorld and 11.0 on WebShop at 7B, reaches target success rates faster, and solves tasks within tighter interaction budgets compared to outcome‑only RL and prior self‑distillation baselines.

arXiv AI
Aug 28

HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents

HINT-SD introduces a targeted self‑distillation framework for long‑horizon language‑model agents that uses full‑trajectory hindsight to identify failure‑relevant actions and applies feedback‑conditioned distillation only to those action spans. This selective approach reduces the need for per‑turn feedback, improving training efficiency and effectiveness. Experiments on BFCL v3 and AppWorld demonstrate that HINT‑SD outperforms dense per‑turn feedback baselines by up to 13.60 percentage points on average while cutting training time per step by 2.26×.

By Woongyeong Yeo, Yumin Choi, Taekyung Ki, Sung Ju Hwang
arXiv AI
Aug 11

Bidirectional Context Self-Distillation for Reinforcement Learning of Skill-Based LLM Agents

arXiv:2608. 09555v1 Announce Type: new Abstract: External natural-language skills provide large language model (LLM) agents with reusable and editable guidance for solving complex tasks.

By Tianjun Pan, Yuan Li, Hongda Wang, Linbo Jin, Mengfei Song, Lei Gao, Qiming Shi, Shaokang Fu, Jiarong Zhao, Chengyu Wang, Chengfu Huo
arXiv AI
Jul 28

From Proprietary to Open-Source: Bridging the Distribution Gap via Multi-Agent Protocol Distillation in Agentic Search

arXiv:2607. 24280v1 Announce Type: new Abstract: 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.

By Junlin Liu, Jiangwang Chen, Zixin Song, Shuaiyu Zhou, Chunji Lv, Hank Wu, Kailin Jiang, Jinyang Wu, Bohan Yu, Chenxi Zhou
arXiv AI
Sep 3

APEx: Distillation of Agent Procedural Experience for Adaptive Deep Research Question Answering

APEx is a hierarchical framework that organizes a deep research agent’s interaction history into instance-level trajectory memories and category-level procedural skills. It couples these through an Executor, Distiller, and Planner, trained with a three-stage alternating GRPO paradigm to enable reward-guided skill distillation. At test time, distilled skills act as procedural priors for online Planner adaptation via skill-guided reinforcement learning, achieving state‑of‑the‑art results on seven benchmarks, outperforming GPT‑5.4 by 14.7 points and the best memory‑augmented baseline by 3.0 points.

By Jie Ding, Rui Sun, Xinyuan Zhang, Zeyu Zhang, Xin Liu