Experiments show that BehR-based training improves long-term alignment in several settings, with the clearest gains in WebShop and less movement in near-ceiling regimes, while preserving or improving single-step prediction quality in three of four settings.
Youling Huang, Guan-Qiao Chen, Junchi Yao et al.· arXiv.org· 4 citations
CoHarden is proposed, a co-generation framework that uses the Lax signal as an in-loop convergence criterion that generates a test before any fix, then iteratively hardens the test and fix against surviving mutation patches until the generated test no longer admits Lax regressions.
Yuhao Tan, Zhibang Yang, Fangkai Yang et al.· arXiv.org· 1 citation
DPIAgent is proposed, a structured agentic framework built on three principles, Divide, Protocol, Isolate (DPI), that mitigates compound-objective ambiguity and goal drift, and shows that architectural structure and backbone capability are complementary axes rather than substitutes, demonstrating DPI's generalizability...
Hao Liu, Steven Liu, Xin Zhang et al.· 0 citations
This work introduces ToolAtlas, a graph-based framework that builds a persistent provider-side tool memory of tool capabilities, failure boundaries, and cross-tool compositions through execution-verified probing and establishes provider-side tool memory as an effective and reusable paradigm for tool servers.
Yue Fang, Zhibang Yang, Fangkai Yang et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
