25 citations · 66 across the 16 of their papers we have counts for
16 papers
Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost
Haojun Xia, Xiaoxia Wu, Jisen Li +12
The KV cache is a dominant memory bottleneck for LLM inference. While 4-bit KV quantization preserves accuracy, 2-bit often degrades it, especially on long-context reasoning. We cl…
Beat the long tail: Distribution-Aware Speculative Decoding for RL Training
Zelei Shao, Vikranth Srivatsa, Sanjana Srivastava +12
Reinforcement learning(RL) post-training has become essential for aligning large language models (LLMs), yet its efficiency is increasingly constrained by the rollout phase, where…
When Does Divide and Conquer Work for Long Context LLM? A Noise Decomposition Framework
Zhen Xu, Shang Zhu, Jue Wang +5
We investigate the challenge of applying Large Language Models (LLMs) to long texts. We propose a theoretical framework that distinguishes the failure modes of long context tasks i…
FloE: On-the-Fly MoE Inference on Memory-constrained GPU
Yuxin Zhou, Zheng Li, Jun Zhang +5
With the widespread adoption of Mixture-of-Experts (MoE) models, there is a growing demand for efficient inference on memory-constrained devices. While offloading expert parameters…
HMI: Hierarchical Knowledge Management for Efficient Multi-Tenant Inference in Pretrained Language Models
Jun Zhang, Jue Wang, Huan Li +6
The significant computational demands of pretrained language models (PLMs), which often require dedicated hardware, present a substantial challenge in serving them efficiently, esp…
Scaling Instruction-Tuned LLMs to Million-Token Contexts via Hierarchical Synthetic Data Generation
Linda He, Jue Wang, Maurice Weber +3
Large Language Models (LLMs) struggle with long-context reasoning, not only due to the quadratic scaling of computational complexity with sequence length but also because of the sc…
Rethinking Inference-Time Scaling: Efficiency Limits and Linguistic Signals
Junlin Wang, Shang Zhu, Jon Saad-Falcon +7
There is intense interest in investigating how inference time compute (ITC) (e.g. repeated sampling, refinements, etc) can improve large language model (LLM) capabilities. While br…
Train Small, Infer Large: Memory-Efficient LoRA Training for Large Language Models
Jun Zhang, Jue Wang, Huan Li +6
Large Language Models (LLMs) have significantly advanced natural language processing with exceptional task generalization capabilities. Low-Rank Adaption (LoRA) offers a cost-effec…
Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping
Muru Zhang, Mayank Mishra, Zhongzhu Zhou +7
Large language model inference is both memory-intensive and time-consuming, often requiring distributed algorithms to efficiently scale. Various model parallelism strategies are us…
Mixture-of-Agents Enhances Large Language Model Capabilities
Junlin Wang, Jue Wang, Ben Athiwaratkun +2
Recent advances in large language models (LLMs) demonstrate substantial capabilities in natural language understanding and generation tasks. With the growing number of LLMs, how to…
Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
Zichang Liu, Jue Wang, Tri Dao +8
Large language models (LLMs) with hundreds of billions of parameters have sparked a new wave of exciting AI applications. However, they are computationally expensive at inference t…
Skill-it! A Data-Driven Skills Framework for Understanding and Training Language Models
Mayee F. Chen, Nicholas Roberts, Kush Bhatia +4
The quality of training data impacts the performance of pre-trained large language models (LMs). Given a fixed budget of tokens, we study how to best select data that leads to good…
Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt
Zhaozhuo Xu, Zirui Liu, Beidi Chen +5
While the numerous parameters in Large Language Models (LLMs) contribute to their superior performance, this massive scale makes them inefficient and memory-hungry. Thus, they are…
Fine-tuning Language Models over Slow Networks using Activation Compression with Guarantees
Jue Wang, Binhang Yuan, Luka Rimanic +5
Communication compression is a crucial technique for modern distributed learning systems to alleviate their communication bottlenecks over slower networks. Despite recent intensive…
Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders
Jue Wang, Wei Lu
Named entity recognition and relation extraction are two important fundamental problems. Joint learning algorithms have been proposed to solve both tasks simultaneously, and many o…
Semi-Supervised Few-Shot Learning for Dual Question-Answer Extraction
Jue Wang, Ke Chen, Lidan Shou +2
This paper addresses the problem of key phrase extraction from sentences. Existing state-of-the-art supervised methods require large amounts of annotated data to achieve good perfo…