Picture for Junyang Lin

Junyang Lin

additional authors not shown

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

Add code
May 10, 2025
Viaarxiv icon

PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts

Add code
Apr 30, 2025
Viaarxiv icon

Qwen2.5-Omni Technical Report

Add code
Mar 26, 2025
Viaarxiv icon

START: Self-taught Reasoner with Tools

Add code
Mar 07, 2025
Viaarxiv icon

InSerter: Speech Instruction Following with Unsupervised Interleaved Pre-training

Add code
Mar 04, 2025
Viaarxiv icon

Multimodal Representation Alignment for Image Generation: Text-Image Interleaved Control Is Easier Than You Think

Add code
Feb 27, 2025
Viaarxiv icon

DataMan: Data Manager for Pre-training Large Language Models

Add code
Feb 26, 2025
Viaarxiv icon

AutoLogi: Automated Generation of Logic Puzzles for Evaluating Reasoning Abilities of Large Language Models

Add code
Feb 24, 2025
Viaarxiv icon

Qwen2.5-VL Technical Report

Add code
Feb 19, 2025
Viaarxiv icon

HellaSwag-Pro: A Large-Scale Bilingual Benchmark for Evaluating the Robustness of LLMs in Commonsense Reasoning

Add code
Feb 17, 2025
Viaarxiv icon