Bairu Hou
YOU?
Author Swipe
View article: WebDART: Dynamic Decomposition and Re-planning for Complex Web Tasks
WebDART: Dynamic Decomposition and Re-planning for Complex Web Tasks Open
Large language model (LLM) agents are becoming competent at straightforward web tasks, such as opening an item page or submitting a form, but still struggle with objectives that require long horizon navigation, large scale information extr…
View article: Research on Accident Type Prediction for New Energy Vehicles Based on the AS-Naive Bayes Algorithm
Research on Accident Type Prediction for New Energy Vehicles Based on the AS-Naive Bayes Algorithm Open
Developing new energy vehicles (NEVs) is a key strategy for achieving low-carbon and sustainable transportation. However, as the number of NEVs increases, traffic accidents involving these vehicles have risen sharply. To explore the charac…
View article: Research on Target Detection Method for Intelligent Mobile Robots based on Machine Visionr
Research on Target Detection Method for Intelligent Mobile Robots based on Machine Visionr Open
Intelligent mobile robots use machine vision and deep learning to achieve environmental perception, path planning, and obstacle avoidance, effectively improving autonomous decision-making capabilities. In this paper, you only look once (YO…
View article: KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse
KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse Open
We describe KVLink, an approach for efficient key-value (KV) cache reuse in large language models (LLMs). In many LLM applications, different inputs can share overlapping context, such as the same retrieved document appearing in multiple q…
View article: Instruction-Following Pruning for Large Language Models
Instruction-Following Pruning for Large Language Models Open
With the rapid scaling of large language models (LLMs), structured pruning has become a widely used technique to learn efficient, smaller models from larger ones, delivering superior performance compared to training similarly sized models …
View article: A Probabilistic Framework for LLM Hallucination Detection via Belief Tree Propagation
A Probabilistic Framework for LLM Hallucination Detection via Belief Tree Propagation Open
This paper focuses on the task of hallucination detection, which aims to determine the truthfulness of LLM-generated statements. To address this problem, a popular class of methods utilize the LLM's self-consistencies in its beliefs in a s…
View article: Advancing the Robustness of Large Language Models through Self-Denoised Smoothing
Advancing the Robustness of Large Language Models through Self-Denoised Smoothing Open
Although large language models (LLMs) have achieved significant success, their vulnerability to adversarial perturbations, including recent jailbreak attacks, has raised considerable concerns. However, the increasing size of these models a…
View article: A Survey on Data Selection for Language Models
A Survey on Data Selection for Language Models Open
A major factor in the recent success of large language models is the use of enormous and ever-growing text datasets for unsupervised pre-training. However, naively training a model on all available data may not be optimal (or feasible), as…
View article: Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing
Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing Open
Aligned large language models (LLMs) are vulnerable to jailbreaking attacks, which bypass the safeguards of targeted LLMs and fool them into generating objectionable content. While initial defenses show promise against token-based threat m…
View article: Decomposing Uncertainty for Large Language Models through Input Clarification Ensembling
Decomposing Uncertainty for Large Language Models through Input Clarification Ensembling Open
Uncertainty decomposition refers to the task of decomposing the total uncertainty of a predictive model into aleatoric (data) uncertainty, resulting from inherent randomness in the data-generating process, and epistemic (model) uncertainty…
View article: Certified Robustness for Large Language Models with Self-Denoising
Certified Robustness for Large Language Models with Self-Denoising Open
Although large language models (LLMs) have achieved great success in vast real-world applications, their vulnerabilities towards noisy inputs have significantly limited their uses, especially in high-stake environments. In these contexts, …
View article: Improving Diffusion Models for Scene Text Editing with Dual Encoders
Improving Diffusion Models for Scene Text Editing with Dual Encoders Open
Scene text editing is a challenging task that involves modifying or inserting specified texts in an image while maintaining its natural and realistic appearance. Most previous approaches to this task rely on style-transfer models that crop…
View article: PromptBoosting: Black-Box Text Classification with Ten Forward Passes
PromptBoosting: Black-Box Text Classification with Ten Forward Passes Open
We describe PromptBoosting, a query-efficient procedure for building a text classifier from a neural language model (LM) without access to the LM's parameters, gradients, or hidden representations. This form of "black-box" classifier train…
View article: TextGrad: Advancing Robustness Evaluation in NLP by Gradient-Driven Optimization
TextGrad: Advancing Robustness Evaluation in NLP by Gradient-Driven Optimization Open
Robustness evaluation against adversarial examples has become increasingly important to unveil the trustworthiness of the prevailing deep models in natural language processing (NLP). However, in contrast to the computer vision domain where…
View article: OpenAttack: An Open-source Textual Adversarial Attack Toolkit
OpenAttack: An Open-source Textual Adversarial Attack Toolkit Open
Textual adversarial attacking has received wide and increasing attention in recent years. Various attack models have been proposed, which are enormously distinct and implemented with different programming frameworks and settings. These fac…
View article: Learning to Attack: Towards Textual Adversarial Attacking in Real-world Situations
Learning to Attack: Towards Textual Adversarial Attacking in Real-world Situations Open
Adversarial attacking aims to fool deep neural networks with adversarial examples. In the field of natural language processing, various textual adversarial attack models have been proposed, varying in the accessibility to the victim model.…
View article: Try to Substitute: An Unsupervised Chinese Word Sense Disambiguation Method Based on HowNet
Try to Substitute: An Unsupervised Chinese Word Sense Disambiguation Method Based on HowNet Open
Word sense disambiguation (WSD) is a fundamental natural language processing task. Unsupervised knowledge-based WSD only relies on a lexical knowledge base as the sense inventory and has wider practical use than supervised WSD that require…