Papers6

#Large Language Models

Rethinking the Trust Region in LLM Reinforcement Learning

Penghui Qi, Xiangxin Zhou et al.Feb 4arXiv

The paper shows that the popular PPO method for training language models is unfair to rare words and too gentle with very common words, which makes learning slow and unstable.

#Reinforcement Learning#Proximal Policy Optimization#Trust Region

KromHC: Manifold-Constrained Hyper-Connections with Kronecker-Product Residual Matrices

Intermediate

Wuyang Zhou, Yuxuan Gu et al.Jan 29arXiv

Hyper-Connections (HC) make the usual single shortcut in neural networks wider by creating several parallel streams and letting the model mix them, but this can become unstable when stacked deep.

#Hyper-Connections#Manifold-Constrained Hyper-Connections#Doubly Stochastic Matrix

AACR-Bench: Evaluating Automatic Code Review with Holistic Repository-Level Context

Intermediate

Lei Zhang, Yongda Yu et al.Jan 27arXiv

AACR-Bench is a new test set that checks how well AI can do code reviews using the whole project, not just one file.

#Automated Code Review#Benchmark#Repository-level Context

Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs

Intermediate

Wei Zhou, Jun Zhou et al.Jan 22arXiv

This survey explains how large language models (LLMs) can clean, connect, and enrich messy data so it’s ready for real apps like dashboards, fraud detection, and training AI.

#Data Preparation#Data Cleaning#Data Integration

MDAgent2: Large Language Model for Code Generation and Knowledge Q&A in Molecular Dynamics

Intermediate

Zhuofan Shi, Hubao A et al.Jan 5arXiv

MDAgent2 is a special helper built from large language models (LLMs) that can both answer questions about molecular dynamics and write runnable LAMMPS simulation code.

#Molecular Dynamics#LAMMPS#Code Generation

mHC: Manifold-Constrained Hyper-Connections

Intermediate

Zhenda Xie, Yixuan Wei et al.Dec 31arXiv

The paper fixes a stability problem in Hyper-Connections (HC) by gently steering the network’s mixing matrix onto a safe shape (a manifold) where signals don’t blow up or vanish.

#Residual Connections#Hyper-Connections#Manifold Projection