Tag
MemSFT is a research paper proposing to mitigate the alignment tax in LLM fine-tuning by using an external parametric memory that decouples domain specialization from backbone parameter updates, enabling reuse across different LLM sizes while preserving general performance.
This paper introduces Confident Decoding, a training-free decoding strategy that dynamically selects the most reliable intermediate layer in LLMs using entropy-guided search, mitigating the alignment tax and improving reasoning performance on benchmarks like GPQA-Diamond and Omni-MATH with negligible overhead.
SDOF is a framework that treats multi-agent execution as a constrained state machine, using an online-RLHF specialized intent router and state-aware dispatcher to enforce business process stage constraints, achieving 86.5% task completion on a recruitment system with 6,000+ enterprises.