THUNLP ORG

Large-scale, Informative, and Diverse Multi-round Chat Data (and Models)

A plug-and-play library for parameter-efficient-tuning (Delta Tuning)

An Open-Source Framework for Prompt-Learning.

An Open-Source Package for Neural Relation Extraction (NRE)

An Open-Source Package for Knowledge Embedding (KE)

An Open-Source Package for Network Embedding (NE)

Must-read papers on graph neural networks (GNN)

A curated reading list of must-read academic papers on graph neural networks, organized into surveys, model architectures, and application areas like knowledge graphs, molecules, social networks, and recommender systems.

An open-source online reverse dictionary.

WantWords is an open-source reverse dictionary: describe a concept in plain English or Chinese and a neural network returns the words that match what you're thinking of.

Must-read papers on prompt-based tuning for pre-trained language models.

Must-read Papers on pre-trained language models.

Must-read papers on network representation learning (NRL) / network embedding (NE)

An Efficient Lexical Analyzer for Chinese

Must-read Papers on Textual Adversarial Attack and Defense

Must-read papers on knowledge representation learning (KRL) / knowledge embedding (KE)

Knowledge Graph Embeddings including TransE, TransH, TransR and PTransE

Source code and dataset for ACL 2019 paper "ERNIE: Enhanced Language Representation with Informative Entities"

THUOCL（THU Open Chinese Lexicon）中文词库

Must-read papers on neural relation extraction (NRE)

Open Chinese Language Pre-trained Model Zoo

No description.

Official codes for ACL 2023 paper "WebCPM: Interactive Web Search for Chinese Long-form Question Answering"

Must-read papers on Machine Reading Comprehension

No description.

An Efficient Lexical Analyzer for Chinese

Neural Relation Extraction, including CNN, PCNN, CNN+ATT, PCNN+ATT

中文谣言数据

An Open-Source Package for Textual Adversarial Attack.

A Large-Scale Few-Shot Relation Extraction Dataset

Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe

Dataset and codes for ACL 2019 DocRED: A Large-Scale Document-Level Relation Extraction Dataset.

Core Data of HowNet and OpenHowNet Python API

A LLM-based Agent that predict its tasks proactively.

An implementation of TransE and its extended models for Knowledge Representation Learning on TensorFlow

Chinese AI & Law Challenge

Must-read Papers on Legal Intelligence

No description.

An Open-Source Package for Information Retrieval.

LLaVA-UHD v3: Progressive Visual Compression for Efficient Native-Resolution Encoding in MLLMs

An Efficient implementation of TransE and its extended models for Knowledge Representation Learning

The code of our paper "InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory"

Code and data of ACL 2021 paper "Few-NERD: A Few-shot Named Entity Recognition Dataset"

No description.

The repo for Tsinghua summer course: Interdisciplinary Seminar on Big Models

Open Platform for Embodied Agents

An Efficient Lexical Analyzer for Chinese

Codes for our paper "ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate"

Survey of Surveys for Natural Language Processing (SOS4NLP)

Neural Sentiment Classification

Must-read Papers of Parameter-Efficient Tuning (Delta Tuning) Methods on Pre-trained Models.

[ICLR 2026 Blogpost Track Poster] JustRL: Scaling a 1.5B LLM with a Simple RL Recipe

Source code for ACL 2019 paper "Chinese Relation Extraction with Multi-Grained Information and External Linguistic Knowledge"

Source code for "Packed Levitated Marker for Entity and Relation Extraction"

An Efficient Chinese Text Classifier

An open-source toolkit for textual backdoor attack and defense (NeurIPS 2022 D&B, Spotlight)

kpt code

The source code of ACL 2018 paper "Denoising Distantly Supervised Open-Domain Question Answering".

Must-read Papers on Sememe Computation

Improved Word Representation Learning with Sememes

Source code and checkpoints for legal pre-trained language models.

Code and dataset of AAAI2019 paper Hybrid Attention-Based Prototypical Networks for Noisy Few-Shot Relation Classification

Joint Neural Relation Extraction with Text and KGs

The source codes for Fine-grained Fact Verification with Kernel Graph Attention Network.

NLP Course Material & QA

Code for Chinese LIWC Lexicon Expansion via Hierarchical Classification of Word Embeddings with Sememe Attention (AAAI18)

OOP Course Material & QA

Prompt Tuning with Rules

No description.

TritonBench: Benchmarking Large Language Model Capabilities for Generating Triton Operators

No description.

The source code of our COLING'18 paper "Few-Shot Charge Prediction with Discriminative Legal Attributes".

THU Chinese Keyphrase Extraction Toolkit

Source code and dataset for ACL2022 Findings Paper "LEVEN: A Large-Scale Chinese Legal Event Detection dataset"

No description.

Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)

No description.

No description.

Source code and dataset for KDD 2020 paper "Adaptive Graph Encoder for Attributed Graph Embedding"

Code and data of the AAAI-20 paper "Multi-channel Reverse Dictionary Model"

Bert-based models(BERT, MTB, CP) for relation extraction.

Source code and datasets of IJCAI2017 paper "TransNet: Translation-Based Network Representation Learning for Social Relation Extraction".

Source code for ACL 2019 paper "GEAR: Graph-based Evidence Aggregating and Reasoning for Fact Verification"

No description.

On Transferability of Prompt Tuning for Natural Language Processing

Hierarchical Neural Relation Extraction

Source code for "A Deep-learning System Bridging Molecule Structure and Biomedical Text with Comprehension Comparable to Human Professionals"

[ACL2025 Findings] Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large Language Models

Dataset and baseline for ACL 2019 paper "XQA: A Cross-lingual Open-domain Question Answering Dataset"

Code and data of the ACL 2020 paper "Word-level Textual Adversarial Attacking as Combinatorial Optimization"

Source code for EMNLP-IJCNLP 2019 paper "HMEAE: Hierarchical Modular Event Argument Extraction".

The repository for paper "DebugBench: "Evaluating Debugging Capability of Large Language Models".

Source code for ACL 2021 paper "ERICA: Improving Entity and Relation Understanding for Pre-trained Language Models via Contrastive Learning"

Representation Learning of Knowledge Graphs with Hierarchical Types (IJCAI-2016)

[ACL'25 Main] ChartCoder: Advancing Multimodal Large Language Model for Chart-to-Code Generation

No description.

Code and data of the EMNLP 2022 paper "Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLP".

Must-read Papers on Neural Information Retrieval

Source code for EMNLP-IJCNLP 2019 paper "Event Detection with Trigger-Aware Lattice Neural Network".

Max-margin DeepWalk

Code for paper "Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System"

Source code for "Train No Evil: Selective Masking for Task-Guided Pre-Training"

KARL: Knowledge-Aware Reasoning and Reinforcement Learning for Knowledge-Intensive Visual Grounding

Code and Data for SIGIR 2020 Paper "Few-Shot Generative Conversational Query Rewriting"

Source code for EMNLP 2020 paper "Coreferential Reasoning Learning for Language Representation"

The official implementation of the paper: H-Neurons: On the Existence, Impact, and Origin of Hallucination-Associated Neurons in LLMs

No description.

The respository of jec-qa.

[ACL 2023] Plug-and-Play Knowledge Injection for Pre-trained Language Models

No description.

Delta-CoMe can achieve near loss-less 1-bit compressin which has been accepted by NeurIPS 2024

Source code for ACL2019 paper "Multi-Channel Graph Neural Network for Entity Alignment".

Evaluate Multimodal LLMs as Embodied Agents

Source code for ACL 2019 paper "DIAG-NRE: A Neural Pattern Diagnosis Framework for Distantly Supervised Neural Relation Extraction".

[ACL 2025 main] FR-Spec: Frequency-Ranked Speculative Sampling

Paragraph Vector Implementation

Source code for EMNLP 2022 paper “PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models”

Revised Version of SAT Model in "Improved Word Representation Learning with Sememes"

Network Representation Learning with Rich Text Information (IJCAI 2015)

Source code for NAACL 2021 paper "TR-BERT: Dynamic Token Reduction for Accelerating BERT Inference"

No description.

Code and data of the EMNLP 2021 paper "Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer"

Code and data of the ACL-IJCNLP 2021 paper "Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger"

No description.

The code and data for ACL2017 paper "Neural Relation Extraction with Multi-lingual Attention"

Image-embodied Knowledge Representation Learning (IJCAI-2017)

Code repo for SIGIR 2021 paper "Few-Shot Conversational Dense Retrieval"

An Efficient Lexical Analyzer for Chinese

Does William Shakespeare REALLY Write Hamlet? Knowledge Representation Learning with Confidence (AAAI-2018)

Source codes of Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction

codes accompanying ACL 2019 paper Quantifying the Similarity between Relations with Fact Distributions

ACDiT: Interpolating Autoregressive Conditional Modeling and Diffusion Transformer

Papers on integrating large language models with embodied AI

Sequence-level 1F1B schedule for LLMs.

Official Implementation of APB (ACL 2025 main Oral) and Spava (ACL 2026 main).

Source code for paper: Knowledge Inheritance for Pre-trained Language Models

Official implementation of the EMNLP 2021 paper "ONION: A Simple and Effective Defense Against Textual Backdoor Attacks"

No description.

Code and models for the paper: Hybrid Linear Attention Done Right: Efficient Distillation and Effective Architectures for Extremely Long Contexts

COS960: A Chinese Word Similarity Dataset of 960 Word Pairs

Source code for EMNLP 2020 paper "Denoising Relation Extraction from Document-level Distant Supervision"

Source code and data for ACL 2019 paper "Modeling Semantic Compositionality with Sememe Knowledge"

No description.

The open-source materials for paper "Sparsing Law: Towards Large Language Models with Greater Activation Sparsity".

Datasets of NCP, contaning news, rumors and legal documents.

source code for TKDE paper “CED: Credible Early Detection of Social Media Rumors”

Code for EMNLP2020 paper "Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment".

CokeBERT: Contextual Knowledge Selection and Embedding towards Enhanced Pre-Trained Language Models

Official repo for "SE-Bench: Benchmarking Self-Evolution with Knowledge Internalization"

Neuron Activation

A High-Performance Lexical Analyzer for Chinese

[ACL 2023 Findings] Emergent Modularity in Pre-trained Transformers

ACL 2024: LoRA-Flow Dynamic LoRA Fusion for Large Language Models in Generative Tasks

Plug-and-Play Document Modules for Pre-trained Models

Code accompanying EMNLP 2018 paper Language Modeling with Sparse Product of Sememe Experts

No description.

Code accompanying Incorporating Chinese Characters of Words for Lexical Sememe Prediction (ACL2018) https://arxiv.org/abs/1806.06349

Official implementation for the paper "KG-Infused RAG: Augmenting Corpus-Based RAG with External Knowledge Graphs"

Learning to Generate STRUCTURED Output with Schema Reinforcement Learning

Code for paper "Beyond Natural Language: LLMs Leveraging Alternative Formats for Enhanced Reasoning and Communication"

Repo for ACL2023 paper "Won't Get Fooled Again: Answering Questions with False Premises"

Source code of paper 'Open Hierarchical Relation Extraction' (NAACL 2021)

Code and data of the AAAI-20 paper "Towards Building a Multilingual Sememe Knowledge Base: Predicting Sememes for BabelNet Synsets"

Source codes for paper "BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity".

Code and data for EMNLP 2018 paper "Cross-lingual Lexical Sememe Prediction"

Neural Entity Typing with Language Model Enhancement

The official implementation of NOSA

No description.

Code and data of the ACL 2021 paper "Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution"

Enhancing Legal Case Retrieval via Scaling High-quality Synthetic Query-Candidate Pairs (EMNLP 2024)

Official codes for COLING 2024 paper "Robust and Scalable Model Editing for Large Language Models": https://arxiv.org/abs/2403.17431v1

Code and data for the COLING 2020 paper "Try to Substitute: An Unsupervised Chinese Word Sense Disambiguation Method Based on HowNet"

No description.

No description.

No description.

Repo for ACL 2024 paper "Tell Me More! Towards Implicit User Intention Understanding of Language Model Driven Agents"

[EMNLP 2025 Findings] ClueAnchor: Clue-Anchored Knowledge Reasoning Exploration and Optimization for Retrieval-Augmented Generation

A large-scale dataset of Chu bamboo slip scripts and a multi-granularity tokenizer for ancient Chinese scripts

A Chinese Character BERT Trained with Multi-Level Masking

KACC: A Multi-task Benchmark for Knowledge Abstraction, Concretization and Completion

CSS-LM: Contrastive Semi-supervised Fine-tuning of Pre-trained Language Models

Open and Knowledgeable NLP Toolkit including CWS, POS Tagging, NER, and Entity Typing

BurstEngine is an efficient framework designed to train LLMs on long-sequence data.

No description.

CA-LoRA: Adapting Existing LoRA for Compressed LLMs to Enable Efficient Multi-Tasking on Personal Devices (COLM 2024)

No description.

Single-Shot Meta-Pruning (SMP) for attention heads of Transformers

[ACL 2025 main] AgentRM: Enhancing Agent Generalization with Reward Modeling

Code and data of the TASLP paper "Improving Sequence Modeling Ability of Recurrent Neural Networks via Sememes"

No description.

Source code for NAACL 2019 paper "Fact Discovery from Knowledge Base via Facet Decomposition".

No description.

No description.

The code for the paper "Cost-Optimal Grouped-Query Attention for Long-Context Modeling"

The official implementation of the paper "StateX: Enhancing RNN Recall via Post-training State Expansion".

Source code for Knowledge Transfer via Pre-training for Recommendation: A Review and Prospect

Rethinking the Role of Efficient Attention in Hybrid Architectures

Source code for paper "DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices".

Official code for "The Overthinker's DIET: Cutting Token Calories with DIfficulty-AwarE Training"

No description.

No description.

The code of the paper Stuffed Mamba: Oversized States Lead to the Inability to Forget