Hybrid Open-Ended Tri-Evolution Makes Better Deep Researcher
Quick Answer
This paper shows that The Hybrid Open-Ended Tri-Evolution (HOTE) framework enhances AI agents' capabilities in open-ended research tasks, outperforming static models.
Quick Take
The Hybrid Open-Ended Tri-Evolution (HOTE) framework enhances AI agents' capabilities in open-ended research tasks, outperforming static models. Experiments show an 8B model trained via HOTE exceeds the performance of 8-32B static models and state-of-the-art methods, with reduced time overhead.
Key Points
- HOTE integrates proposer, solver, and judge for collaborative evolution.
- 8B model trained via HOTE outperforms 8-32B static models.
- HOTE reduces time overhead compared to traditional deep research methods.
- Extensive experiments validate the necessity of evolving all three HOTE modules.
- Framework aims to advance autonomous agent capabilities in open-ended environments.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 13710v1 Announce Type: new Abstract: Deep research and agent evolution serve as de-facto tasks for AI agents in real-world applications toward artificial general intelligence. The former enables autonomous retrieval and integration of information in open-ended environments to tackle open-ended research tasks, yet it is constrained by the static parametric deep research capabilities of agent systems.
The latter allows agents to autonomously interact with the environment to gain experiences that evolve model capabilities. However, its effectiveness has been widely validated only on verifiable tasks with standard answers, leaving a gap with open-ended research tasks.
To bridge these two critical tasks, we propose the Hybrid Open-Ended Tri-Evolution (HOTE) framework, which leverages hybrid-mode reinforcement learning to facilitate the collaborative evolution of a proposer, solver and judge based on web-scale knowledge, moving toward autonomous evolving agents in open-ended tasks and environments.
Extensive experiments on three long-form deep research benchmarks demonstrate that the 8B model trained via HOTE surpasses the strongest static open 8-32B models as well as those trained by state-of-the-art deep research training methods with less time overhead, and further verify that the evolution of all three modules in HOTE is indispensable.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →Arbor: Tree Search as a Cognition Layer for Autonomous Agents
Arbor introduces a multi-agent framework utilizing structured tree search for optimizing LLM inference, achieving up to 193% throughput-latency improvement compared to vendor-optimized systems. It employs an Orchestrator and Critic agent for stability and coordination, demonstrating hardware-agnostic performance with minimal variance.