This work revisits the forgetting target of LLM unlearning and introduces Knowledge Association (KA), which characterizes the association between a query context and its target knowledge and is instantiated by the corresponding conditional predictive distribution, as the fundamental target of forgetting.
Wu-Tong Xu, Yun Qu, Hai-Yan Feng et al.· 0 citations
TreeOPSD is introduced, a critic-free, tree-aware posterior self-distillation framework for long-horizon agentic reinforcement learning that improves agentic capabilities across several competitive benchmarks and model scales, especially for long-horizon tasks.
Heming Zou, Qi Wang, Yi-Xiu Mao et al.· 2 citations
This work formalizes failure modes, derives lagged-risk decompositions that expose the resulting temporal bias, and recasts risk prediction as an in-context sequence modeling problem, and yields more reliable difficulty estimates and consistent performance gains across robust Meta-RL, DR, and prompt curriculum RL post-...
Jia-Rong Wen, Qing-Chao Tao, Kai-Yu Zhang et al.· 1 citation
In-context learning (ICL) is crucial for boosting the inference performance of large language models (LLMs). However, the effectiveness of ICL in LLMs is greatly influenced by the choice of demonstration sets. Exhaustive searches over these sets are combinatorial, and existing selectors often rely on relevance or likel...