11 papers
DataFoundry: Evolving Data Preparators via Recursive Self-Improvement
Cehao Yang, Xiaojun Wu, Xueyuan Lin +4
Domain adaptation of large language models increasingly depends on constructing high-quality training data, yet existing data-preparation pipelines typically address quality only a…
Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning
Lu Dai, Ziyang Rao, Yili Wang +3
Fine-tuning LLMs to inject new knowledge faces a critical challenge: LLMs can quickly memorize new facts, yet fail to use them for downstream reasoning tasks. We formalize this fai…
LLM-Oriented Information Retrieval: A Denoising-First Perspective
Lu Dai, Liang Sun, Fanpu Cao +4
Modern information retrieval (IR) is no longer consumed primarily by humans but increasingly by large language models (LLMs) via retrieval-augmented generation (RAG) and agentic se…
When Looking Is Not Enough: Visual Attention Structure Reveals Hallucination in MLLMs
Fanpu Cao, Xin Zou, Xuming Hu +1
Multimodal large language models (MLLMs) have become a key interface for visual reasoning and grounded question answering, yet they remain vulnerable to visual hallucinations, wher…
SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing
Mingxu Zhang, Yuhan Li, Lujundong Li +3
Large language models possess strong chemical reasoning capabilities, making them effective molecular editors. However, property-relevant information is implicitly entangled across…
TableVision: A Large-Scale Benchmark for Spatially Grounded Reasoning over Complex Hierarchical Tables
Xiaoyu Chen, Lu Dai, Hanqing Wang +6
Structured tables are essential for conveying high-density information in professional domains such as finance, healthcare, and scientific research. Despite the progress in Multimo…
RETuning: Upgrading Inference-Time Scaling for Stock Movement Prediction with Large Language Models
Xueyuan Lin, Cehao Yang, Ye Ma +7
Recently, large language models (LLMs) have demonstrated outstanding reasoning capabilities on mathematical and coding tasks. However, their application to financial tasks-especial…
GraphSearch: An Agentic Deep Searching Workflow for Graph Retrieval-Augmented Generation
Cehao Yang, Xiaojun Wu, Xueyuan Lin +6
Graph Retrieval-Augmented Generation (GraphRAG) enhances factual reasoning in LLMs by structurally modeling knowledge through graph-based representations. However, existing GraphRA…
Think-on-Graph 3.0: Efficient and Adaptive LLM Reasoning on Heterogeneous Graphs via Multi-Agent Dual-Evolving Context Retrieval
Xiaojun Wu, Cehao Yang, Xueyuan Lin +6
Graph-based Retrieval-Augmented Generation (GraphRAG) has become the important paradigm for enhancing Large Language Models (LLMs) with external knowledge. However, existing approa…
Select2Reason: Efficient Instruction-Tuning Data Selection for Long-CoT Reasoning
Cehao Yang, Xueyuan Lin, Xiaojun Wu +5
A practical approach to activate long chain-of-thoughts reasoning ability in pre-trained large language models is to perform supervised fine-tuning on instruction datasets synthesi…
LongFaith: Enhancing Long-Context Reasoning in LLMs with Faithful Synthetic Data
Cehao Yang, Xueyuan Lin, Chengjin Xu +5
Despite the growing development of long-context large language models (LLMs), data-centric approaches relying on synthetic data have been hindered by issues related to faithfulness…