Conghui He

Lost in Tokenization: Context as the Key to Unlocking Biomolecular Understanding in Scientific LLMs

Earth-Agent: Unlocking the Full Landscape of Earth Observation with Agents

Dripper: Token-Efficient Main HTML Extraction with a Lightweight LM

The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs

Not All Documents Are What You Need for Extracting Instruction Tuning Data

Reinforcement Learning for Evidence-Seeking Diagnostic Reasoning with Large Language Models

IDEAL: Data Equilibrium Adaptation for Multi-Capability Language Model Alignment

Rethinking LLM Evaluation: Can We Evaluate LLMs with 200× Less Data?

SAS-Bench: A Fine-Grained Benchmark for Evaluating Short Answer Scoring with Large Language Models

OPV: Outcome-based Process Verifier for Efficient Long Chain-of-Thought Verification

FLARE: Fully Integration of Vision-Language Representations for Deep Cross-Modal Understanding

ScaleDiff: Scaling Difficult Problems for Advanced Mathematical Reasoning

Data-efficient Online Training for Direct Alignment in LLMs

AudioMarathon: A Comprehensive Benchmark for Long-Context Audio Understanding and Efficiency in Audio LLMs

OmniLayout: Enabling Coarse-to-Fine Learning with LLMs for Universal Document Layout Generation

Spot the Fake: Large Multimodal Model-Based Synthetic Image Detection with Artifact Explanation

LOKI: A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Models

OmniCorpus: A Unified Multimodal Corpus of 10 Billion-Level Images Interleaved with Text

GeoX: Geometric Problem Solving Through Unified Formalized Vision-Language Pre-training

Scaling Code-Assisted Chain-of-Thoughts and Instructions for Model Reasoning

Multi-step Visual Reasoning with Visual Tokens Scaling and Verification

Harnessing Diversity for Important Data Selection in Pretraining Large Language Models

Hallucination at a Glance: Controlled Visual Edits and Fine-Grained Multimodal Learning

Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluation

DocGenome: A Large Benchmark for Multi-Modal Language Models in Real-World Academic Document Understanding

Efficient Multi-modal Large Language Models via Progressive Consistency Distillation

MIA-DPO: Multi-Image Augmented Direct Preference Optimization For Large Vision-Language Models

Multi-Agent Collaborative Data Selection for Efficient Language Model Pretraining

Unearthing Large Scale Domain-Specific Knowledge from Public Corpora

IPDreamer: Appearance-Controllable 3D Object Generation with Complex Image Prompts

DocLayout-YOLO: Enhancing Document Layout Analysis through Diverse Synthetic Data and Global-to-Local Adaptive Perception

SongComposer: A Large Language Model for Lyric and Melody Composition in Song Generation

Trustworthy Dataset Proof: Certifying the Authentic Use of Dataset in Training Models for Enhanced Trust

BEATS: Optimizing LLM Mathematical Capabilities with BackVerify and Adaptive Disambiguate based Efficient Tree Search

CMIRA: Class Membership Inducing Recovery Attacks Against Machine Unlearning Models

Gradual Learning: Optimizing Fine-Tuning with Partially Mastered Knowledge in Large Language Models

PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction