Dimensions overall score 8.0

• Chenxi Wang* 1 2 Zhuoyun Yu* 1 2 Xin Xie 2 Wuguannan Yao2 Runnan Fang 1 Shuofei Qiao 1 Kexin Cao 1 Guozhou Zheng 1 Xiang Qi 2 Peng Zhang 2 Shumin Deng 1 Abstract Learning from experience is critical for building capable large language model (LLM) agentsclaim 72%
• be the tasks set. We write q∈ Qfor sampling a task, and let R(τ, q)∈ {0,1}be a task-dependent success indicator. We model the LLM agent as a policy π that induces a trajectory distribution. Without external skills, the agent generates trajectories by direct reasoning: τ∼π(· |q), q∈ Qclaim 72%
• et al., 2025a). Among them, GLM-4.6 has been reported to exhibit strong native agentic capabilities in agent mid- training, serving as a competitive backbone for our study. We compare against four representative baselines: (1) No- memory, which performs inference without retrieving any prior experienceclaim 72%
• times per training task, followed by skill extraction, skill refinement, and skill expansion. The maximum number of refinement iterations is set to 3. For efficiency, we limit environment exploration to one rollout per training task; the sampling temperature is 1.0 during explorationclaim 72%
• ExpeL∗ 57.33 77.67 32.87 58.93 56.25 42.50 39.38 ExpeL‡ 59.33 78.83 32.94 58.78 58.12 43.75 41.25 SkillX‡ 63.67 82.00 35.12 58.93 66.87 47.50 43.75 Kimi-K2-Instruct-0905 No Memory∗ 65.17 78.00 46.88 70.24 75.62 51.25 78.12 A-Mem∗ 65.17 76.67 46.58 72.62 76.25 52.50 76.87claim 72%
• ExpeL∗ 66.33 79.33 52.53 78.57 77.50 55.50 78.75 ExpeL‡ 66.00 79.67 52.98 78.87 77.50 56.25 79.37 SkillX‡ 66.83 81.33 56.40 81.55 78.12 58.75 82.50claim 72%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card