Skip to content
@LUMIA-Group

LUMIA: Language Understanding and Machine Intelligence Algorithms

LUMIA is a research group lead by Zhouhan Lin in Shanghai Jiao Tong University.

Popular repositories Loading

  1. MemoryDecoder MemoryDecoder Public

    The official implementation of the paper "Memory Decoder: A Pretrained, Plug-and-Play Memory for Large Language Models" (NeurIPS 2025 Poster).

    Python 70 7

  2. Leveraging-Self-Supervised-Learning-for-AVSR Leveraging-Self-Supervised-Learning-for-AVSR Public

    Official PyTorch implementation of paper Leveraging Unimodal Self Supervised Learning for Multimodal Audio-Visual Speech Recognition (ACL 2022)

    Python 68 13

  3. rasat rasat Public

    Forked from ServiceNow/picard

    The official implementation of the paper "RASAT: Integrating Relational Structures into Pretrained Seq2Seq Model for Text-to-SQL"(EMNLP 2022)

    Python 64 17

  4. MLPMemory MLPMemory Public

    The official implementation of the paper "MLP Memory: A Retriever-Pretrained Memory for Large Language Models". (ICLR 2026)

    Python 52 2

  5. OrderedGNN OrderedGNN Public

    The official implementation of the paper "Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing" (ICLR 2023).

    Python 48 5

  6. FourierTransformer FourierTransformer Public

    The official Pytorch implementation of the paper "Fourier Transformer: Fast Long Range Modeling by Removing Sequence Redundancy with FFT Operator" (ACL 2023 Findings)

    Python 40 1

Repositories

Showing 10 of 29 repositories
  • AWM Public

    Official implementation of the paper "AWM: Accurate Weight-Matrix Fingerprint for Large Language Models"

    LUMIA-Group/AWM’s past year of commit activity
    Python 9 1 0 0 Updated Feb 20, 2026
  • ConceptLM Public

    Official Implementation of ConceptLM.

    LUMIA-Group/ConceptLM’s past year of commit activity
    Python 12 0 0 0 Updated Feb 5, 2026
  • FreqKV Public

    The official implementation of the paper "FreqKV: Key-Value Compression in Frequency Domain for Context Window Extension" (ICLR 2026).

    LUMIA-Group/FreqKV’s past year of commit activity
    Python 3 0 0 0 Updated Jan 30, 2026
  • MLPMemory Public

    The official implementation of the paper "MLP Memory: A Retriever-Pretrained Memory for Large Language Models". (ICLR 2026)

    LUMIA-Group/MLPMemory’s past year of commit activity
    Python 52 Apache-2.0 2 1 0 Updated Jan 28, 2026
  • PonderLM-2 Public

    Official implementation of the paper "PonderLM-2: Pretraining LLM with Latent Thoughts in Continuous Space"

    LUMIA-Group/PonderLM-2’s past year of commit activity
    Python 7 1 1 0 Updated Oct 24, 2025
  • ContextLM Public
    LUMIA-Group/ContextLM’s past year of commit activity
    Python 12 1 2 0 Updated Oct 16, 2025
  • MemoryDecoder Public

    The official implementation of the paper "Memory Decoder: A Pretrained, Plug-and-Play Memory for Large Language Models" (NeurIPS 2025 Poster).

    LUMIA-Group/MemoryDecoder’s past year of commit activity
    Python 70 Apache-2.0 7 2 0 Updated Sep 29, 2025
  • Anchor-Embedding Public

    The official implementation of the paper "Training LLMs to be Better Text Embedders through Bidirectional Reconstruction" (EMNLP 2025 Main Conference).

    LUMIA-Group/Anchor-Embedding’s past year of commit activity
    Python 9 1 0 0 Updated Aug 28, 2025
  • TosT5 Public

    The official implementation of the paper "Enhancing SPARQL Generation by Triplet-order-sensitive Pre-training" (CIKM 2024).

    LUMIA-Group/TosT5’s past year of commit activity
    Python 3 MIT 1 0 0 Updated Aug 27, 2025
  • PonderingLM Public Forked from hiyouga/LlamaFactory

    Official implementation of the paper "Pretraining Language Models to Ponder in Continuous Space"

    LUMIA-Group/PonderingLM’s past year of commit activity
    Python 25 Apache-2.0 8,806 0 0 Updated Jul 21, 2025

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Python Vue

Most used topics

Loading…