papers

Publications (11)

cs.LG2026

What Makes Position Zero Special? A Mechanistic Study of Position Zero Attention Sinks in LLMs

Runyu Peng, Ruixiao Li, Mingshu Chen +5

Transformers frequently allocate disproportionate attention to specific tokens, a phenomenon known as attention sinks. Causal large language models reliably form one at position ze…

cs.LG2026

Explicit Multi-head Attention for Inter-head Interaction in Large Language Models

Runyu Peng, Yunhua Zhou, Demin Song +4

In large language models built upon the Transformer architecture, recent studies have shown that inter-head interaction can enhance attention performance. Motivated by this, we pro…

cs.LG2025

Intern-S1: A Scientific Multimodal Foundation Model

Lei Bai, Zhongrui Cai, Yuhang Cao +173

In recent years, a plethora of open-source foundation models have emerged, achieving remarkable progress in some widely attended fields, with performance being quite close to that…

cs.CL2025

Data-free Weight Compress and Denoise for Large Language Models

Runyu Peng, Yunhua Zhou, Qipeng Guo +4

Large Language Models (LLMs) are reshaping the research landscape in artificial intelligence, particularly as model parameters scale up significantly, unlocking remarkable capabili…

cs.AI2026

How to Set the Batch Size for Large-Scale Pre-training?

Yunhua Zhou, Junhao Huang, Shuhao Xing +4

The concept of Critical Batch Size, as pioneered by OpenAI, has long served as a foundational principle for large-scale pre-training. However, with the paradigm shift towards the W…

cs.CL2025

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

Qin Zhu, Fei Huang, Runyu Peng +6

While logical reasoning evaluation of Large Language Models (LLMs) has attracted significant attention, existing benchmarks predominantly rely on multiple-choice formats that are v…

cs.LG2025

Implicit Reward as the Bridge: A Unified View of SFT and DPO Connections

Bo Wang, Qinyuan Cheng, Runyu Peng +7

Post-training processes are essential phases in grounding pre-trained language models to real-world tasks, with learning from demonstrations or preference signals playing a crucial…

cs.CL2024

WanJuan-CC: A Safe and High-Quality Open-sourced English Webtext Dataset

Jiantao Qiu, Haijun Lv, Zhenjiang Jin +23

This paper presents WanJuan-CC, a safe and high-quality open-sourced English webtext dataset derived from Common Crawl data. The study addresses the challenges of constructing larg…

cs.LG2026

Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale

Yicheng Zou, Dongsheng Zhu, Lin Zhu +174

We introduce Intern-S1-Pro, the first one-trillion-parameter scientific multimodal foundation model. Scaling to this unprecedented size, the model delivers a comprehensive enhancem…

cs.CL2024

Inference-Time Decontamination: Reusing Leaked Benchmarks for Large Language Model Evaluation

Qin Zhu, Qingyuan Cheng, Runyu Peng +5

The training process of large language models (LLMs) often involves varying degrees of test data contamination. Although current LLMs are achieving increasingly better performance…

cs.CL2026

Synthetic Pre-Pre-Training Improves Language Model Robustness to Noisy Pre-Training Data

Xu Guo, Runyu Peng, Jian Tong +4

Large language models (LLMs) rely on web-scale corpora for pre-training. The noise inherent in these datasets tends to obscure meaningful patterns and ultimately degrade model perf…