paper-with-me

Papers

ABSeeker: Training Long-Horizon Search Agents via Answer-Backtracked Credit Assignment

2026-08-05 · Yijun Lu, Rui Ye, Jiajun Wang, Yuwen Du, Tian Jin, Songhua Liu, Siheng Chen hf

Long-horizon search agents must make multiple sequential actions (steps) to search, retrieve, verify, and integrate evidence to reach a final answer. However, existing methods for training these agents typically treat all steps within a trajectory uniformly during both supervised fine-tuning (SFT) and reinforcement learning (RL), failing to distinguish useful actions from erroneous or redundant ones. In this paper, we propose Answer-Backtracked Credit Assignment (ABC), a fine-grained credit assignment framework for training long-horizon search agents by converting sparse trajectory-level outcomes into dense step-level supervision that rewards useful actions (even in failed trajectories) while suppressing erroneous or redundant actions. Specifically, given a potentially obscure query and its corresponding ground-truth answer, ABC first performs Answer-Backtracked Clue Recovery, which traces back from the answer to recover intermediate clues required to solve the question. It then applies Clue-Anchored Step Scoring to evaluate each search step against these clues, converting sparse binary outcome supervision into dense step-level rewards. Based on these rewards, we develop ABC-SFT, which reweights the loss of each turn, and ABC-GRPO, which uses the step-level scores as rewards in GRPO. Building on this framework, we train ABSeeker based on Qwen3.5-4B with only 8.5k examples. ABSeeker achieves 37.3% on BrowseComp and 39.1% on BrowseComp-ZH. With context management, the scores further improve to 55.3% and 52.9%, respectively, significantly outperforming same-scale (4B) agents and even matching the performance of larger ones (approximately 30B). These results demonstrate the effectiveness of answer-backtracked step-level credit assignment for training long-horizon search agents.

📄 PDF Abstract BibTeX arXiv:2608.05102

Code (4)

Aaron617/agent-arXiv-daily ★ 10
PolarSeeker/ABSeeker ★ 47
Tavish9/awesome-daily-AI-arxiv ★ 113
🤗 PolarSeeker/ABSeeker-4B-RL ★ 3

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agents

2026-03-29 · Zhaopeng Feng, Liangcai Su, Zhen Zhang, Xinyu Wang 외 arxiv

As large language models (LLMs) evolve into autonomous agents for long-horizon information-seeking, managing finite context capacity has become a critical bottleneck. Existing context management methods typically commit …

Reinforcement Learning for Long-Horizon Multi-Turn Search Agents

2025-10-28 · Vivek Kalyan, Martin Andrews arxiv

Large Language Model (LLM) agents can leverage multiple turns and tools to solve complex tasks, with prompt-based approaches achieving strong performance. This work demonstrates that Reinforcement Learning (RL) can push …

Reinforcement Learning

IterResearch: Rethinking Long-Horizon Agents with Interaction Scaling

2025-11-10 · Guoxin Chen, Zile Qiao, Xuanzhong Chen, Donglei Yu 외 arxiv

Recent advances in deep-research agents have shown promise for autonomous knowledge construction through dynamic reasoning over external sources. However, existing approaches rely on a mono-contextual paradigm that accum…

DeepSearch-World: Self-Distillation for Deep Search Agents in a Verifiable Environment

2026-07-08 · Xinyu Geng, Xuanhua He, Sixiang Chen, Yanjing Xiao 외 arxiv

Training tool-use agents to improve from their own experience remains challenging, as supervised fine-tuning relies on fixed teacher-distilled trajectories, while sparse-reward reinforcement learning provides weak superv…

Reinforcement Learning

Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RL

2025-08-11 · Jiaxuan Gao, Wei Fu, Minyang Xie, Shusheng Xu 외 arxiv

Recent advancements in LLM-based agents have demonstrated remarkable capabilities in handling complex, knowledge-intensive tasks by integrating external tools. Among diverse choices of tools, search tools play a pivotal …