Picture for Xiaodong Yu

Xiaodong Yu

Zeroth-Order Fine-Tuning of LLMs with Extreme Sparsity

Jun 05, 2024
Viaarxiv icon

Lightweight Spatial Modeling for Combinatorial Information Extraction From Documents

Add code
May 08, 2024
Viaarxiv icon

Automatic Hallucination Assessment for Aligned Large Language Models via Transferable Adversarial Attacks

Add code
Oct 19, 2023
Figure 1 for Automatic Hallucination Assessment for Aligned Large Language Models via Transferable Adversarial Attacks
Figure 2 for Automatic Hallucination Assessment for Aligned Large Language Models via Transferable Adversarial Attacks
Figure 3 for Automatic Hallucination Assessment for Aligned Large Language Models via Transferable Adversarial Attacks
Figure 4 for Automatic Hallucination Assessment for Aligned Large Language Models via Transferable Adversarial Attacks
Viaarxiv icon

Benchmarking and In-depth Performance Study of Large Language Models on Habana Gaudi Processors

Sep 29, 2023
Viaarxiv icon

Building Interpretable and Reliable Open Information Retriever for New Domains Overnight

Add code
Aug 09, 2023
Figure 1 for Building Interpretable and Reliable Open Information Retriever for New Domains Overnight
Figure 2 for Building Interpretable and Reliable Open Information Retriever for New Domains Overnight
Figure 3 for Building Interpretable and Reliable Open Information Retriever for New Domains Overnight
Figure 4 for Building Interpretable and Reliable Open Information Retriever for New Domains Overnight
Viaarxiv icon

HEAT: A Highly Efficient and Affordable Training System for Collaborative Filtering Based Recommendation on CPUs

Add code
May 03, 2023
Figure 1 for HEAT: A Highly Efficient and Affordable Training System for Collaborative Filtering Based Recommendation on CPUs
Figure 2 for HEAT: A Highly Efficient and Affordable Training System for Collaborative Filtering Based Recommendation on CPUs
Figure 3 for HEAT: A Highly Efficient and Affordable Training System for Collaborative Filtering Based Recommendation on CPUs
Figure 4 for HEAT: A Highly Efficient and Affordable Training System for Collaborative Filtering Based Recommendation on CPUs
Viaarxiv icon

SOLAR: A Highly Optimized Data Loading Framework for Distributed Training of CNN-based Scientific Surrogates

Add code
Nov 04, 2022
Figure 1 for SOLAR: A Highly Optimized Data Loading Framework for Distributed Training of CNN-based Scientific Surrogates
Figure 2 for SOLAR: A Highly Optimized Data Loading Framework for Distributed Training of CNN-based Scientific Surrogates
Figure 3 for SOLAR: A Highly Optimized Data Loading Framework for Distributed Training of CNN-based Scientific Surrogates
Figure 4 for SOLAR: A Highly Optimized Data Loading Framework for Distributed Training of CNN-based Scientific Surrogates
Viaarxiv icon

Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts

Oct 30, 2022
Figure 1 for Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
Figure 2 for Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
Figure 3 for Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
Figure 4 for Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
Viaarxiv icon

SOIT: Segmenting Objects with Instance-Aware Transformers

Add code
Dec 23, 2021
Figure 1 for SOIT: Segmenting Objects with Instance-Aware Transformers
Figure 2 for SOIT: Segmenting Objects with Instance-Aware Transformers
Figure 3 for SOIT: Segmenting Objects with Instance-Aware Transformers
Figure 4 for SOIT: Segmenting Objects with Instance-Aware Transformers
Viaarxiv icon

Event Linking: Grounding Event Mentions to Wikipedia

Add code
Dec 15, 2021
Figure 1 for Event Linking: Grounding Event Mentions to Wikipedia
Figure 2 for Event Linking: Grounding Event Mentions to Wikipedia
Figure 3 for Event Linking: Grounding Event Mentions to Wikipedia
Figure 4 for Event Linking: Grounding Event Mentions to Wikipedia
Viaarxiv icon