arXiv Machine Learning By Estelle Zheng, Nathan Cerisara, S\'ebastien Warichet, Emmanuel Helbert, Christophe Cerisara

Ladder Up, Memory Down: Low-Cost Fine-Tuning With Side Nets

Read the original on arXiv Machine Learning →

The paper introduces Ladder Side Tuning (LST), a parameter‑efficient fine‑tuning method that adds a lightweight side network to large language models. LST matches QLoRA’s compute scaling while halving peak memory usage, enabling 7B‑parameter models to be fine‑tuned on a single 12 GB GPU with 2k‑token contexts without gradient checkpointing. The authors also present xLadder, a depth‑extended variant that increases effective depth through cross‑connections, allowing deeper reasoning without extra memory overhead.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Hugging Face Trending Papers
Aug 18

PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX

PTXBench is a benchmark designed to evaluate and adapt large language models (LLMs) for GPU kernel optimization using architecture-specific PTX code. It assesses functional correctness, runtime execution of target instructions, and speedup over leading libraries on GEMM and attention workloads on H100 and B200 GPUs. The study finds uneven performance across models, especially on complex attention backward tasks, and shows that fine‑tuning Qwen3.6‑27B improves some tasks but generalization remains inconsistent.

arXiv AI
Aug 19

PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX

PTXBench is a benchmark designed to evaluate and adapt large language models (LLMs) for GPU kernel optimization using architecture‑specific PTX code. It assesses functional correctness, runtime execution of target instructions, and speedup over state‑of‑the‑art libraries on GEMM and attention workloads on H100 and B200 GPUs. The study finds uneven success rates, especially on complex attention backward tasks, and shows that executing target instructions does not guarantee competitive performance, with no model consistently outperforming frontier libraries. The authors also fine‑tune Qwen3.6‑27B, noting that repair‑conditioned training improves some tasks but generalization remains inconsistent, highlighting the importance of data coverage, balance, and teacher quality.

By Genghan Zhang, Yixin Dong, Chengze Fan, Zhichen Zeng, Yueming Yuan, Shaowei Zhu, Kunle Olukotun
arXiv AI
Aug 18

AutoMem: A Text-Gradient Recursive Self-Improvement Framework for Automated Memory Architectures Search

arXiv:2608. 14621v1 Announce Type: cross Abstract: Long-term memory is increasingly central to LLM agents, yet memory design remains a highly coupled architecture problem: what to encode, how to store it, how to retrieve it, and how to manage it can vary substantially across tasks and backbone models.

By Lin Du, Jie Zhou, Yuxuan Cai, Kai Chen, Qin Chen, Xin Li, Bo Zhang, Wei Li, Liang He
arXiv AI
Aug 18

SAPE: Sandwich Adapters for Parameter Efficiency in Large Language Model Fine-Tuning

arXiv:2608. 15360v1 Announce Type: cross Abstract: While Parameter-Efficient Fine-Tuning (PEFT) has substantially reduced the hardware cost of adapting Large Language Models (LLMs) by decreasing the number of trainable parameters, recent studies have sought to further improve PEFT through parameter sharing.

By Mohammad Aref Jafari-Raddani, Morteza Mohajjel Kafshdooz