System Message Generation for User Preferences using Open-Source Models

• URL: http://arxiv.org/abs/2502.11330v2
• Date: Fri, 23 May 2025 01:00:16 GMT
• Title: System Message Generation for User Preferences using Open-Source Models
• Authors: Minbyul Jeong, Jungho Cho, Minsoo Khang, Dawoon Jung, Teakgyu Hong,
• Abstract summary: System messages play a crucial role in interactions with large language models (LLMs)<n>We introduce SysGen, a pipeline for generating system messages that better align assistant responses with user instructions.<n>Training open-source models on SysGen data yields substantial improvements in both single-turn (Multifacet) and multi-turn (SysBench) conversation benchmarks.
• Score: 4.387048445855714
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: System messages play a crucial role in interactions with large language models (LLMs), often serving as prompts to initiate conversations. Through system messages, users can assign specific roles, perform intended tasks, incorporate background information, and specify various output formats and communication styles. Despite such versatility, publicly available datasets often lack system messages and are subject to strict license constraints in industrial applications. Moreover, manually annotating system messages that align with user instructions is resource-intensive. In light of these challenges, we introduce SysGen, a pipeline for generating system messages that better align assistant responses with user instructions using existing supervised fine-tuning datasets that lack system messages. Training open-source models on SysGen data yields substantial improvements in both single-turn (Multifacet) and multi-turn (SysBench) conversation benchmarks. Notably, our method shows strong gains in shorter conversations, suggesting that it enhances early-stage interaction effectiveness. Our qualitative analysis further emphasizes the value of diverse and structured system messages in improving LLM adaptability across varied user scenarios.

Related papers

• A New Paradigm of User-Centric Wireless Communication Driven by Large Language Models [53.16213723669751]
  Next generation of wireless communications seeks to deeply integrate artificial intelligence with user-centric communication networks.<n>We propose a novel paradigm for wireless communication that innovatively incorporates the nature language to structured query language.<n>We present a prototype system in which a dynamic semantic representation network at the physical layer adapts its encoding depth to meet user requirements.
  arXiv  Detail & Related papers  (2025-04-16T01:43:36Z)
• Benchmarking Retrieval-Augmented Generation in Multi-Modal Contexts [56.30364248231053]
  This paper introduces Multi-Modal Retrieval-Augmented Generation (M2RAG) M2RAG is a benchmark designed to evaluate the effectiveness of Multi-modal Large Language Models (MLLMs) To enhance the context utilization capabilities of MLLMs, we also introduce Multi-Modal Retrieval-Augmented Instruction Tuning (MM-RAIT)
  arXiv  Detail & Related papers  (2025-02-24T16:25:25Z)
• TALKPLAY: Multimodal Music Recommendation with Large Language Models [6.830154140450626]
  We present TALKPLAY, a novel multimodal music recommendation system that reformulates recommendation as a token generation problem using large language models (LLMs)<n>Our system effectively recommends music from diverse user queries while generating contextually relevant responses.<n>Our qualitative and quantitative evaluation demonstrates that TALKPLAY significantly outperforms unimodal approaches based solely on text or listening history in both recommendation performance and conversational naturalness.
  arXiv  Detail & Related papers  (2025-02-19T13:28:20Z)
• Has My System Prompt Been Used? Large Language Model Prompt Membership Inference [56.20586932251531]
  We develop Prompt Detective, a statistical method to reliably determine whether a given system prompt was used by a third-party language model.<n>Our work reveals that even minor changes in system prompts manifest in distinct response distributions, enabling us to verify prompt usage with statistical significance.
  arXiv  Detail & Related papers  (2025-02-14T08:00:42Z)
• SysBench: Can Large Language Models Follow System Messages? [30.701602680394686]
  Large Language Models (LLMs) have become instrumental across various applications, with the customization of these models to specific scenarios becoming increasingly critical. Despite the recognized potential of system messages to optimize AI-driven solutions, there is a notable absence of a benchmark for evaluating how well LLMs follow system messages. We introduce SysBench, a benchmark that systematically analyzes system message following ability in terms of three limitations of existing LLMs.
  arXiv  Detail & Related papers  (2024-08-20T15:33:16Z)
• Training Zero-Shot Generalizable End-to-End Task-Oriented Dialog System Without Turn-level Dialog Annotations [2.757798192967912]
  This work employs multi-task instruction fine-tuning to create more efficient and scalable task-oriented dialogue systems. Our approach outperforms both state-of-the-art models trained on annotated data and billion-scale parameter off-the-shelf ChatGPT models.
  arXiv  Detail & Related papers  (2024-07-21T04:52:38Z)
• Multi-modal Instruction Tuned LLMs with Fine-grained Visual Perception [63.03288425612792]
  We propose bfAnyRef, a general MLLM model that can generate pixel-wise object perceptions and natural language descriptions from multi-modality references. Our model achieves state-of-the-art results across multiple benchmarks, including diverse modality referring segmentation and region-level referring expression generation.
  arXiv  Detail & Related papers  (2024-03-05T13:45:46Z)
• UniMS-RAG: A Unified Multi-source Retrieval-Augmented Generation for Personalized Dialogue Systems [43.266153244137215]
  Large Language Models (LLMs) has shown exceptional capabilities in many natual language understanding and generation tasks. We decompose the use of multiple sources in generating personalized response into three sub-tasks: Knowledge Source Selection, Knowledge Retrieval, and Response Generation. We propose a novel Unified Multi-Source Retrieval-Augmented Generation system (UniMS-RAG)
  arXiv  Detail & Related papers  (2024-01-24T06:50:20Z)
• DialCLIP: Empowering CLIP as Multi-Modal Dialog Retriever [83.33209603041013]
  We propose a parameter-efficient prompt-tuning method named DialCLIP for multi-modal dialog retrieval. Our approach introduces a multi-modal context generator to learn context features which are distilled into prompts within the pre-trained vision-language model CLIP. To facilitate various types of retrieval, we also design multiple experts to learn mappings from CLIP outputs to multi-modal representation space.
  arXiv  Detail & Related papers  (2024-01-02T07:40:12Z)
• Multi-User MultiWOZ: Task-Oriented Dialogues among Multiple Users [51.34484827552774]
  We release the Multi-User MultiWOZ dataset: task-oriented dialogues among two users and one agent. These dialogues reflect interesting dynamics of collaborative decision-making in task-oriented scenarios. We propose a novel task of multi-user contextual query rewriting: to rewrite a task-oriented chat between two users as a concise task-oriented query.
  arXiv  Detail & Related papers  (2023-10-31T14:12:07Z)
• Talk the Walk: Synthetic Data Generation for Conversational Music Recommendation [62.019437228000776]
  We present TalkWalk, which generates realistic high-quality conversational data by leveraging encoded expertise in widely available item collections. We generate over one million diverse conversations in a human-collected dataset.
  arXiv  Detail & Related papers  (2023-01-27T01:54:16Z)
• Pivotal Role of Language Modeling in Recommender Systems: Enriching Task-specific and Task-agnostic Representation Learning [23.119223101680976]
  We show that language modeling applied directly to task-specific user histories achieves excellent results on diverse recommendation tasks. We further demonstrate that our approach can provide promising transfer learning capabilities for a broad spectrum of real-world recommender systems.
  arXiv  Detail & Related papers  (2022-12-07T16:31:14Z)
• Building a Role Specified Open-Domain Dialogue System Leveraging Large-Scale Language Models [15.062014096238803]
  We study the challenge of imposing roles on open-domain dialogue systems. We propose an efficient data collection framework for building role-satisfying dialogue dataset from scratch. Our models return few out-of-bounds utterances, keeping competitive performance on general metrics.
  arXiv  Detail & Related papers  (2022-04-30T06:23:06Z)
• Disentangling Online Chats with DAG-Structured LSTMs [55.33014148383343]
  DAG-LSTMs are a generalization of Tree-LSTMs that can handle directed acyclic dependencies. We show that the novel model we propose achieves state of the art status on the task of recovering reply-to relations.
  arXiv  Detail & Related papers  (2021-06-16T18:00:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.