124 citations · 158 across the 22 of their papers we have counts for
23 papers
AstraFlow: Dataflow-Oriented Reinforcement Learning for Agentic LLMs
Haizhong Zheng, Yizhuo Di, Jiahui Wang +7
Reinforcement learning (RL) is increasingly used to improve the reasoning, coding, and tool-use capabilities of large language models, but agentic RL remains prohibitively expensiv…
WMVLM: Evaluating Diffusion Model Image Watermarking via Vision-Language Models
Zijin Yang, Yu Sun, Kejiang Chen +4
Digital watermarking is essential for securing generated images from diffusion models. Accurate watermark evaluation is critical for algorithm development, yet existing methods hav…
The Path Not Taken: RLVR Provably Learns Off the Principals
Hanqing Zhu, Zhenyu Zhang, Hanxian Huang +11
Reinforcement Learning with Verifiable Rewards (RLVR) reliably improves the reasoning performance of large language models, yet it appears to modify only a small fraction of parame…
Prosperity before Collapse: How Far Can Off-Policy RL Reach with Stale Data on LLMs?
Haizhong Zheng, Jiawei Zhao, Beidi Chen
Reinforcement learning has been central to recent advances in large language model reasoning, but most algorithms rely on on-policy training that demands fresh rollouts at every up…
PSRT: Accelerating LRM-based Guard Models via Prefilled Safe Reasoning Traces
Jiawei Zhao, Yuang Qi, Weiming Zhang +2
Large Reasoning Models (LRMs) have demonstrated remarkable performance on tasks such as mathematics and code generation. Motivated by these strengths, recent work has empirically d…
Deep Think with Confidence
Yichao Fu, Xuewei Wang, Yuandong Tian +1
Large Language Models (LLMs) have shown great potential in reasoning tasks through test-time scaling methods like self-consistency with majority voting. However, this approach ofte…
Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts
Haizhong Zheng, Yang Zhou, Brian R. Bartoldson +4
Reinforcement learning, such as PPO and GRPO, has powered recent breakthroughs in LLM reasoning. Scaling rollout to sample more prompts enables models to selectively use higher-qua…
GaLore 2: Large-Scale LLM Pre-Training by Gradient Low-Rank Projection
DiJia Su, Andrew Gu, Jane Xu +2
Large language models (LLMs) have revolutionized natural language understanding and generation but face significant memory bottlenecks during training. GaLore, Gradient Low-Rank Pr…
HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading
Cheng Luo, Zefan Cai, Hanshi Sun +7
Transformer-based large language models (LLMs) demonstrate impressive performance in long context generation. Extending the context length has disproportionately shifted the memory…
ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization
Zechun Liu, Changsheng Zhao, Hanxian Huang +13
The optimal bit-width for achieving the best trade-off between quantized model size and accuracy has been a subject of ongoing debate. While some advocate for 4-bit quantization, o…
TensorGRaD: Tensor Gradient Robust Decomposition for Memory-Efficient Neural Operator Training
Sebastian Loeschcke, David Pitt, Robert Joseph George +5
Scientific problems require resolving multi-scale phenomena across different resolutions and learning solution operators in infinite-dimensional function spaces. Neural operators p…
SFT: Efficient, Scalable and Generalizable LLM Fine-tuning by Structured Sparsity
Xinyu Yang, Jixuan Leng, Geyang Guo +5
Current PEFT methods for LLMs can achieve either high quality, efficient training, or scalable serving, but not all three simultaneously. To address this limitation, we investigate…
SQL Injection Jailbreak: A Structural Disaster of Large Language Models
Jiawei Zhao, Kejiang Chen, Weiming Zhang +1
Large Language Models (LLMs) are susceptible to jailbreak attacks that can induce them to generate harmful content. Previous jailbreak methods primarily exploited the internal prop…
Prefix Guidance: A Steering Wheel for Large Language Models to Defend Against Jailbreak Attacks
Jiawei Zhao, Kejiang Chen, Xiaojian Yuan +1
In recent years, the rapid development of large language models (LLMs) has achieved remarkable performance across various tasks. However, research indicates that LLMs are vulnerabl…
Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients
Zhenyu Zhang, Ajay Jaiswal, Lu Yin +4
Training Large Language Models (LLMs) is memory-intensive due to the large number of parameters and associated optimization states. GaLore, a recent method, reduces memory usage by…
Mini-Sequence Transformer: Optimizing Intermediate Memory for Long Sequences Training
Cheng Luo, Jiawei Zhao, Zhuoming Chen +2
We introduce Mini-Sequence Transformer (MsT), a simple and effective methodology for highly efficient and accurate LLM training with extremely long sequences. MsT partitions input…
From Low Rank Gradient Subspace Stabilization to Low-Rank Weights: Observations, Theories, and Applications
Ajay Jaiswal, Yifan Wang, Lu Yin +6
Large Language Models' (LLMs) weight matrices can often be expressed in low-rank form with potential to relax memory and compute resource requirements. Unlike prior efforts that fo…
GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection
Jiawei Zhao, Zhenyu Zhang, Beidi Chen +3
Training Large Language Models (LLMs) presents significant memory challenges, predominantly due to the growing size of weights and optimizer states. Common memory-reduction approac…
Silent Guardian: Protecting Text from Malicious Exploitation by Large Language Models
Jiawei Zhao, Kejiang Chen, Xiaojian Yuan +3
The rapid development of large language models (LLMs) has yielded impressive success in various downstream tasks. However, the vast potential and remarkable capabilities of LLMs al…
ZerO Initialization: Initializing Neural Networks with only Zeros and Ones
Jiawei Zhao, Florian Schäfer, Anima Anandkumar
Deep neural networks are usually initialized with random weights, with adequately selected initial variance to ensure stable signal propagation during training. However, selecting…
LNS-Madam: Low-Precision Training in Logarithmic Number System using Multiplicative Weight Update
Jiawei Zhao, Steve Dai, Rangharajan Venkatesan +6
Representing deep neural networks (DNNs) in low-precision is a promising approach to enable efficient acceleration and memory reduction. Previous methods that train DNNs in low-pre…
A Review of Computer Vision Methods in Network Security
Jiawei Zhao, Rahat Masood, Suranga Seneviratne
Network security has become an area of significant importance more than ever as highlighted by the eye-opening numbers of data breaches, attacks on critical infrastructure, and mal…
signSGD with Majority Vote is Communication Efficient And Fault Tolerant
Jeremy Bernstein, Jiawei Zhao, Kamyar Azizzadenesheli +1
Training neural networks on large datasets can be accelerated by distributing the workload over a network of machines. As datasets grow ever larger, networks of hundreds or thousan…