Tag
This paper presents the first comprehensive empirical study of safety impacts of benign multilingual fine-tuning on LLMs, showing that safety outcomes vary drastically by language and that assessing only English is insufficient.
Proposes FoLoRA, a forgetting-aware optimization framework for fine-tuning foundation models that balances task utility and forgetting penalty via generalized Rayleigh-quotient optimization, achieving better preservation of non-target capabilities.
This paper introduces GLoRA, a gauge-aware server representation for Federated LoRA that addresses the semantic mismatch in factor aggregation by estimating a consensus update subspace. Experiments show GLoRA outperforms baselines in performance and efficiency across heterogeneous client scenarios.
This article recommends a video that systematically explains the 10 core papers shaping today's AI industry, covering Transformer, LoRA, RAG, Agents, and the MCP protocol, aiming to help engineers clarify the technological lineage.