Efficient Trigger Word Insertion
The article, ‘Efficient Trigger Word Insertion’, focuses on the threats posed by backdoor attacks on deep neural network models in the field of Natural Language Processing (NLP). The authors propose…
Continue readingComputation and Language
This focuses on computational approaches to understanding, processing, and generating human languages.
ProAgent: From Robotic Process Automation to Agentic Process Automation
This paper presents a new automation paradigm, Agentic Process Automation (APA), which uses Large Language Models (LLMs) to offload human labor in both the construction and execution of tasks. The…
Continue readingGPT-4V(ision) for Robotics: Multimodal Task Planning from Human Demonstration
The paper introduces a pipeline that enhances a Vision Language Model, GPT-4V(ision), by incorporating human action observations to facilitate robotic manipulation. This system analyzes videos of humans performing tasks and…
Continue readingLabeled Interactive Topic Models
The article ‘Labeled Interactive Topic Models’ by Kyle Seelman, Mozhi Zhang, and Jordan Boyd-Graber discusses the enhancement of neural topic models through user interactivity. The authors correct the lack of…
Continue readingReducing Privacy Risks in Online Self-Disclosures with Language Models
This article discusses a study aimed at protecting user-side privacy in online self-disclosure. The researchers developed a taxonomy of 19 self-disclosure categories and used a language model for identification, achieving…
Continue readingOn the Exploitability of Reinforcement Learning with Human Feedback for Large Language Models
This academic article discusses the security vulnerabilities in Reinforcement Learning with Human Feedback (RLHF) in Large Language Models (LLMs). RLHF plays a crucial role in aligning LLMs with human preferences….
Continue readingAligning with Whom? Large Language Models Have Gender and Racial Biases in Subjective NLP Tasks
The study investigates the biases in large language models (LLMs) when handling subjective natural language processing (NLP) tasks. Using the POPQUORN dataset, the study tests LLMs on their understanding of…
Continue readingIs A Helpful Assistant the Best Role for Large Language Models? A Systematic Evaluation of Social Roles in System Prompts
The study presents a systematic evaluation of how social roles in system prompts influence the performance of large language models (LLMs). The researchers curated a list of 162 roles spanning…
Continue readingGPT-4V(ision) as A Social Media Analysis Engine
This paper investigates the potential of GPT-4V in analyzing social media content. The study involves five tasks: sentiment analysis, hate speech detection, fake news identification, demographic inference, and political ideology…
Continue readingSPHINX: The Joint Mixing of Weights, Tasks, and Visual Embeddings for Multi-modal Large Language Models
The article presents SPHINX, a versatile multi-modal large language model (MLLM) that incorporates a joint mixing strategy of model weights, tuning tasks, and visual embeddings. The model is designed to…
Continue reading