8 papers
KVCapsule: Efficient Sequential KV Cache Compression for Vision-Language Models with Asymmetric Redundancy
Yingbing Huang, Tharun Adithya Srikrishnan, Steven K. Reinhardt +1
Vision-Language Models (VLMs) have emerged as a critical and fast-growing extension of Large Language Models (LLMs) that enable multimodal reasoning through both text and image inp…
Can Asymmetric Tile Buffering Be Beneficial?
Chengyue Wang, Wesley Pang, Xinrui Wu +9
General matrix multiplication (GEMM) is the computational backbone of modern AI workloads, and its efficiency is critically dependent on effective tiling strategies. Conventional a…
TB or Not TB: Coverage-Driven Direct Preference Optimization for Verilog Stimulus Generation
Bardia Nadimi, Khashayar Filom, Deming Chen +1
With the rapid advancement of Large Language Models (LLMs), there is growing interest in applying them to hardware design and verification. Among these stages, design verification…
SchemaCoder: Automatic Log Schema Extraction Coder with Residual Q-Tree Boosting
Lily Jiaxin Wan, Chia-Tung Ho, Rongjian Liang +3
Log schema extraction is the process of deriving human-readable templates from massive volumes of log data, which is essential yet notoriously labor-intensive. Recent studies have…
SpecMemo: Speculative Decoding is in Your Pocket
Selin Yildirim, Deming Chen
Recent advancements in speculative decoding have demonstrated considerable speedup across a wide array of large language model (LLM) tasks. Speculative decoding inherently relies o…
Building Machine Learning Challenges for Anomaly Detection in Science
Elizabeth G. Campolongo, Yuan-Tang Chou, Ekaterina Govorkova +148
Scientific discoveries are often made by finding a pattern or object that was not predicted by the known rules of science. Oftentimes, these anomalous events or objects that do not…
JAM: Controllable and Responsible Text Generation via Causal Reasoning and Latent Vector Manipulation
Yingbing Huang, Deming Chen, Abhishek K. Umrawal
While large language models (LLMs) have made significant strides in generating coherent and contextually relevant text, they often function as opaque black boxes, trained on vast u…
On the Surprising Efficacy of Distillation as an Alternative to Pre-Training Small Models
Sean Farhat, Deming Chen
In this paper, we propose that small models may not need to absorb the cost of pre-training to reap its benefits. Instead, they can capitalize on the astonishing results achieved b…