6 papers
Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation
Pingzhi Tang, Yiding Wang, Muhan Zhang
Large Language Models (LLMs) face the "knowledge cutoff" challenge, where their frozen parametric memory prevents direct internalization of new information. While Supervised Fine-T…
Breaking the Blocks: Continuous Low-Rank Decomposed Scaling for Unified LLM Quantization and Adaptation
Pingzhi Tang, Ruijie Zhou, Fanxu Meng +2
Current quantization methods for LLMs predominantly rely on block-wise structures to maintain efficiency, often at the cost of representational flexibility. In this work, we demons…
HD-PiSSA: High-Rank Distributed Orthogonal Adaptation
Yiding Wang, Fauxu Meng, Xuefeng Zhang +3
Existing parameter-efficient fine-tuning (PEFT) methods for large language models (LLMs), such as LoRA and PiSSA, constrain model updates to low-rank subspaces, limiting their expr…
TPLA: Tensor Parallel Latent Attention for Efficient Disaggregated Prefill and Decode Inference
Xiaojuan Tang, Fanxu Meng, Pingzhi Tang +4
Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, compresses key-value states into a low-rank latent vector, caching only this vector to reduce memory. In tensor parall…
TransMLA: Multi-Head Latent Attention Is All You Need
Fanxu Meng, Pingzhi Tang, Xiaojuan Tang +3
In this paper, we present TransMLA, a framework that seamlessly converts any GQA-based pre-trained model into an MLA-based model. Our approach enables direct compatibility with Dee…
CLOVER: Cross-Layer Orthogonal Vectors Pruning and Fine-Tuning
Fanxu Meng, Pingzhi Tang, Fan jiang +1
Decoder-only models generate tokens autoregressively by caching key/value vectors, but as the cache grows, inference becomes memory-bound. To address this issue, we introduce CLOVE…