Latest AI News

arXiv cs.CL•about 16 hours ago·5 min read

GroupGPT: A Token-efficient and Privacy-preserving Agentic Framework for Multi-User Chat Assistant

arXiv:2603.01059v3 Announce Type: replace Abstract: Recent advances in large language models (LLMs) have enabled increasingly capable chatbots. However, most existing systems focus on single-user settings and do not generalize well to multi-user group chat interactions, where agents require more proactive and accurate intervention under complex, evolving contexts. Existing approaches typically rely on LLMs for both intervention reasoning and response generation, leading to high token consumption, limited scalability, and potential privacy risks. To address these challenges, we propose GroupGPT, a token-efficient and privacy-preserving agentic framework for multi-user chat assistant. GroupGPT adopts an edge-cloud model collaboration architecture to decouple intervention timing from response generation, enabling efficient and accurate decision-making while preserving user privacy through on-device processing of sensitive information. The framework also supports multimodal inputs, including memes, images, videos, and voice messages.To support evaluation of timing accuracy and response quality, we further introduce MUIR, a benchmark dataset for multi-user chat assistant intervention reasoning. MUIR contains 2,500 annotated group chat segments with intervention labels and rationales. We evaluate a range of models on MUIR, spanning from open-source to proprietary variants, including both LLMs and their smaller counterparts. Extensive experiments demonstrate that GroupGPT generates accurate and well-timed responses, achieving an average score of 4.72/5.0 in LLM-based evaluation, and is well-received by users across diverse group chat scenarios. Moreover, GroupGPT reduces the token usage by up to 3 times compared to baselines, while providing privacy sanitization of user messages before cloud transmission. Code is available at: https://github.com/Eliot-Shen/GroupGPT .

Latest News

Luwen Technical Report

Blending Human and LLM Expertise to Detect Hallucinations and Omissions in Mental Health Chatbot Responses

Latest News

Luwen Technical Report

Blending Human and LLM Expertise to Detect Hallucinations and Omissions in Mental Health Chatbot Responses

Understanding Structured Financial Data with LLMs: A Case Study on Fraud Detection

LLM-Based Data Generation and Clinical Skills Evaluation for Low-Resource French OSCEs

Transforming the Voice of the Customer: Large Language Models for Identifying Customer Needs

Quantum Vision Theory Applied to Audio Classification for Deepfake Speech Detection

HiCI: Hierarchical Construction-Integration for Long-Context Attention

Efficient Provably Secure Linguistic Steganography via Range Coding

Contextual Earnings-22: A Speech Recognition Benchmark with Custom Vocabulary in the Wild

HumanLLM: Benchmarking and Improving LLM Anthropomorphism via Human Cognitive Patterns

ORACLE-SWE: Quantifying the Contribution of Oracle Information Signals on SWE Agents

Contrastive Decoding Mitigates Score Range Bias in LLM-as-a-Judge

GroupGPT: A Token-efficient and Privacy-preserving Agentic Framework for Multi-User Chat Assistant

Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Diagnostic Reasoning

Automating Database-Native Function Code Synthesis with LLMs

The Human Condition as Reflected in Contemporary Large Language Models

Compact Example-Based Explanations for Language Models

A GAN and LLM-Driven Data Augmentation Framework for Dynamic Linguistic Pattern Modeling in Chinese Sarcasm Detection

STRIDE-ED: A Strategy-Grounded Stepwise Reasoning Framework for Empathetic Dialogue Systems

EventWeave: A Dynamic Framework for Capturing Core and Supporting Events in Dialogue Systems