This is the official implementation of "Relational In-Context Learning via Synthetic Pre-training with Structural Prior"

Design hardware-friendly model architectures and migrate existing LLMs with minimal performance loss

The official repository of "LIFT: A Novel Framework for Enhancing Long-Context Understanding of LLMs via Long Input Fine-Tuning".

The official repository of "LooGLE v2: Are LLMs Ready for Real World Long Dependency Challenges?" (NeurIPS DB Track 2025)

official code for "Beyond Single-Task: Robust Multi-Task Length Generalization for LLMs" (NeurIPS 2025)

The official repository of The Road Less Traveled: Enhancing Exploration in LLMs via Sequential Sampling.

[EMNLP 2025] HD-PiSSA: High-Rank Distributed Orthogonal Adaptation

PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language Models(NeurIPS 2024 Spotlight)