Source code for our paper ''Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling''

[WISA 2025] This is the code repo for our WISA'25 paper "Adapting Language Models to Text Matching based Recommendation Systems".

Long-Context LLM Preference Optimization

This is the code repo for the paper "Learning to Route Queries Across Knowledge Bases for Step-wise Retrieval-Augmented Reasoning".

[SIGIR 2025] This is the code repo for our SIGIR'25 paper: Enhancing the Patent Matching Capability of Large Language Models via Memory Graph.

Codes for the paper: KBAlign - Efficient Self Adaptation on Specific Knowledge Bases

Source code for our paper: Denoising Sequential Recommendation through User-Consistent Preference Modeling.

Source code for paper "ExpandR: Teaching Dense Retrievers Beyond Queries with LLM Guidance"

Source code for our paper ''RankCoT: Refining Knowledge for Retrieval-Augmented Generation through Ranking Chain-of-Thoughts''

HIPPO: Enhancing the Table Understanding Capability of Large Language Models through Hybrid-Modal Preference Optimization

This is the code repo for our paper "Benchmarking Retrieval-Augmented Generation in Multi-Modal Contexts".

ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation

This is the code repo for our paper "Learning More Effective Representations for Dense Retrieval through Deliberate Thinking Before Search".

Data and software for building the ACL Anthology.

[ADMA 2025] This is the code repo for our ADMA'25 paper: LegalDuet: Learning Fine-grained Representations for Legal Judgment Prediction via a Dual-View Contrastive Learning.

UltraRAG 2.0: Less Code, Lower Barrier, Faster Deployment! MCP-based low-code RAG framework, enabling researchers to build complex pipelines to creative innovation.

This is the code repo for the paper "RAG-DDR: Optimizing Retrieval-Augmented Generation Using Differentiable Data Rewards".

[ICLR 2025] This is the code repo for our ICLR’25 paper "RAG-DDR: Optimizing Retrieval-Augmented Generation Using Differentiable Data Rewards".

Parsing-free RAG supported by VLMs

[TBD 2024] This is the code repo for our TBD‘2024 paper "Multi-Evidence based Fact Verification via A Confidential Graph Neural Network".

This is the code repo for our paper "Language Memory Can Aid Unsupervised Fact Error Correction".

使用pytorch实现简易的神经网络模型的训练的模板，包括并不仅限于Resnet，AlexNet，Agg16，预训练模型等等