Two exact federated unlearning algorithms are proposed, Bi-Models Training (bmt) and Multi-Models Training (mmt), that ensure immediate exact unlearning by design while achieving better performance post-unlearning.
Ze-Yu Zhang, N. Bui, A. Verma et al.· Machine-mediated learning· 1 citation
Agent harnesses enable large language models to act through tools, observe the results, and iteratively work toward a task goal. Verification checks and revises agents'work to improve output quality. Yet there is limited guidance on how harnesses should organize this effort. We study verification as an axis of test-tim...
Zheng-Xin Zhang, Er-Kang Zhu, Yin Lin et al.· 0 citations
A unified theory for RE(S) is developed that covers the full spectrum of S, and can be interpreted as a stage-wise optimization process, where each stage takes $S$ gradient steps for minimizing the Kullback-Leibler distance to a fixed reward-weighted rollout distribution.
Zhi-Wei Wang, Yan-Xi Chen, Ya-Liang Li et al.· 0 citations
This work presents a general framework for training large language models (LLMs) to "Connect the Dots", a meta-capability required by long-lifecycle agents, and presents proof-of-concept implementations of the CoD framework, including a GRPO-style RL algorithm with fine-grained credit assignment.
Yan-Xi Chen, Wei-Jie Shi, Bo Hu et al.· arXiv.org· 0 citations
This paper formalizes preference privacy, a label-DP-style privacy notion for DPO that protects only the relative preference between candidate responses, assuming an adversary who already knows the prompt and responses, and designs PrivDPO, a DPO variant that enforces preference privacy while remaining compatible with...
Yang-Fan Jiang, Fei Wei, Ergute Bao et al.· 0 citations
LLM agents increasingly take on long-running tasks whose history grows far beyond a single model context window. Existing approaches compress earlier interactions or extract selected information into fixed memory representations, committing to what to preserve before future needs are known. We present Scroll, a context...
EvoSOP is introduced, a framework that empowers agents to extract SOPs from execution trajectories and iteratively optimize the toolset through a systematic lifecycle of construction, merging, evaluation, and pruning, providing a scalable pathway for the development of self-evolving agents.
Haipeng Ding, Yuexiang Xie, Zhewei Wei et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.