WebAggregator: Enhancing Compositional Reasoning Capabilities of Deep Research Agent Foundation Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917444378951680 |
|---|---|
| author | Wang, Rui Zhang, Ce Ma, Jun-Yu Zhang, Jianshu Wang, Hongru Chen, Yi Xue, Boyang Fang, Tianqing Zhang, Zhisong Zhang, Hongming Mi, Haitao Yu, Dong Wong, Kam-Fai |
| author_facet | Wang, Rui Zhang, Ce Ma, Jun-Yu Zhang, Jianshu Wang, Hongru Chen, Yi Xue, Boyang Fang, Tianqing Zhang, Zhisong Zhang, Hongming Mi, Haitao Yu, Dong Wong, Kam-Fai |
| contents | The hallmark of Deep Research agents lies in compositional reasoning, the capacity to aggregate distributed, heterogeneous information into coherent logical insights. However, current agentic systems are often retrieval-heavy but reasoning-light, where success is predominantly determined by simple entity-seeking rather than the multi-step aggregation of scattered evidence. To address this, we propose a data synthesis pipeline WebAggregator, designed to shift the agentic paradigm from retrieval-centric to compositional aggregation. Our approach first employs Proactive Explorer to collect interconnected knowledge, then Compositional Logic Proposer to weave knowledge into complex questions using over 12 composition guidelines derived from a rigorous deconstruction of the Deep Research problem setting. By leveraging 10K verifiable QA pairs grounded on 50K websites, we curate a high-quality SFT dataset via rejection sampling. Fine-tuning on this corpus fundamentally transforms agent behavior, fostering deliberate composition reasoning and reduced tool redundancy. The resulting WebAggregator-32B surpasses GPT-4.1 and matches Claude-3.7-Sonnet on GAIA, WebWalkerQA, and XBench. To address the lack of benchmarks that emphasize both reasoning and retrieval, we introduce the WebAggregatorQA testbed, which reveals that even with perfect retrieval, top-tier models still underperformed. These results demonstrate that compositional reasoning, not retrieval, is the true performance ceiling for next-generation research agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_14438 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WebAggregator: Enhancing Compositional Reasoning Capabilities of Deep Research Agent Foundation Models Wang, Rui Zhang, Ce Ma, Jun-Yu Zhang, Jianshu Wang, Hongru Chen, Yi Xue, Boyang Fang, Tianqing Zhang, Zhisong Zhang, Hongming Mi, Haitao Yu, Dong Wong, Kam-Fai Computation and Language The hallmark of Deep Research agents lies in compositional reasoning, the capacity to aggregate distributed, heterogeneous information into coherent logical insights. However, current agentic systems are often retrieval-heavy but reasoning-light, where success is predominantly determined by simple entity-seeking rather than the multi-step aggregation of scattered evidence. To address this, we propose a data synthesis pipeline WebAggregator, designed to shift the agentic paradigm from retrieval-centric to compositional aggregation. Our approach first employs Proactive Explorer to collect interconnected knowledge, then Compositional Logic Proposer to weave knowledge into complex questions using over 12 composition guidelines derived from a rigorous deconstruction of the Deep Research problem setting. By leveraging 10K verifiable QA pairs grounded on 50K websites, we curate a high-quality SFT dataset via rejection sampling. Fine-tuning on this corpus fundamentally transforms agent behavior, fostering deliberate composition reasoning and reduced tool redundancy. The resulting WebAggregator-32B surpasses GPT-4.1 and matches Claude-3.7-Sonnet on GAIA, WebWalkerQA, and XBench. To address the lack of benchmarks that emphasize both reasoning and retrieval, we introduce the WebAggregatorQA testbed, which reveals that even with perfect retrieval, top-tier models still underperformed. These results demonstrate that compositional reasoning, not retrieval, is the true performance ceiling for next-generation research agents. |
| title | WebAggregator: Enhancing Compositional Reasoning Capabilities of Deep Research Agent Foundation Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.14438 |