Fine-tuning on simulated data outperforms prompting for agent tone of voice

• URL: http://arxiv.org/abs/2507.04889v1
• Date: Mon, 07 Jul 2025 11:23:20 GMT
• Title: Fine-tuning on simulated data outperforms prompting for agent tone of voice
• Authors: Ingo Marquardt, Philippe Brule,
• Abstract summary: This study investigates the effectiveness of fine-tuning versus system prompting for aligning language models with a specific behavioral target.<n>Our results demonstrate that fine-tuning outperformed system prompting, achieving a high percentage of conversational responses.<n>We conclude that fine-tuning small, open-weights LMs on simulated data is a highly effective and data-efficient method for instilling specific stylistic behaviors.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deploying language models (LMs) in customer-facing speech applications requires conversational fluency and adherence to specific stylistic guidelines. This can be challenging to achieve reliably using complex system prompts due to issues like instruction following limitations and in-context bias. This study investigates the effectiveness of fine-tuning versus system prompting for aligning LMs with a specific behavioral target: responding in a natural, conversational tone suitable for voice interactions. We fine-tuned a small, open-weights model (`Llama3.2-1B-Instruct`) using Low-Rank Adaptation (LoRA) on a synthetically generated dataset derived from Wikipedia. Additionally, we fine-tuned two closed-source models (`gpt-4o-mini`, `gpt-4.1-mini`). Our results demonstrate that fine-tuning outperformed system prompting, achieving a high percentage of conversational responses, even when trained on only 100 data samples. Semantic similarity analysis confirmed that fine-tuning did not degrade content quality. Interestingly, fine-tuning with 8-bit integer quantization converged faster towards the target style than using bfloat16 precision, potentially due to implicit regularization effects. We conclude that fine-tuning small, open-weights LMs on simulated data is a highly effective and data-efficient method for instilling specific stylistic behaviors, offering a preferable alternative to complex system prompting for practical applications requiring nuanced response styles.

Related papers

• Learning LLM Preference over Intra-Dialogue Pairs: A Framework for Utterance-level Understandings [9.763273544617176]
  Large language models (LLMs) have demonstrated remarkable capabilities in handling complex dialogue tasks without requiring use case-specific fine-tuning.<n>In this paper, we introduce a simple yet effective framework to address this challenge.<n>Our approach is specifically designed for per-utterance classification problems, which encompass tasks such as intent detection, dialogue state tracking, and more.
  arXiv  Detail & Related papers  (2025-03-07T17:46:13Z)
• SPARC: Score Prompting and Adaptive Fusion for Zero-Shot Multi-Label Recognition in Vision-Language Models [74.40683913645731]
  Zero-shot multi-label recognition (MLR) with Vision-Language Models (VLMs) faces significant challenges without training data, model tuning, or architectural modifications.<n>Our work proposes a novel solution treating VLMs as black boxes, leveraging scores without training data or ground truth.<n>Analysis of these prompt scores reveals VLM biases and AND''/OR' signal ambiguities, notably that maximum scores are surprisingly suboptimal compared to second-highest scores.
  arXiv  Detail & Related papers  (2025-02-24T07:15:05Z)
• Towards Generalizable Trajectory Prediction Using Dual-Level Representation Learning And Adaptive Prompting [107.4034346788744]
  Existing vehicle trajectory prediction models struggle with generalizability, prediction uncertainties, and handling complex interactions.<n>We propose Perceiver with Register queries (PerReg+), a novel trajectory prediction framework that introduces: (1) Dual-Level Representation Learning via Self-Distillation (SD) and Masked Reconstruction (MR), capturing global context and fine-grained details; (2) Enhanced Multimodality using register-based queries and pretraining, eliminating the need for clustering and suppression; and (3) Adaptive Prompt Tuning during fine-tuning, freezing the main architecture and optimizing a small number of prompts for efficient adaptation.
  arXiv  Detail & Related papers  (2025-01-08T20:11:09Z)
• CoT-based Synthesizer: Enhancing LLM Performance through Answer Synthesis [31.953858122298517]
  We propose a novel inference scaling strategy, CoT-based Synthesizer.<n>It synthesizes superior answers by analyzing complementary information from multiple candidate responses.<n>We show that our method significantly enhances performance, with gains of 11.8% for Llama3-8B and 10.3% for GPT-4o.
  arXiv  Detail & Related papers  (2025-01-03T06:50:06Z)
• A Systematic Examination of Preference Learning through the Lens of Instruction-Following [83.71180850955679]
  We use a novel synthetic data generation pipeline to generate 48,000 instruction unique-following prompts.<n>With our synthetic prompts, we use two preference dataset curation methods - rejection sampling (RS) and Monte Carlo Tree Search (MCTS)<n>Experiments reveal that shared prefixes in preference pairs, as generated by MCTS, provide marginal but consistent improvements.<n>High-contrast preference pairs generally outperform low-contrast pairs; however, combining both often yields the best performance.
  arXiv  Detail & Related papers  (2024-12-18T15:38:39Z)
• ToolACE: Winning the Points of LLM Function Calling [139.07157814653638]
  ToolACE is an automatic agentic pipeline designed to generate accurate, complex, and diverse tool-learning data.<n>We demonstrate that models trained on our synthesized data, even with only 8B parameters, achieve state-of-the-art performance on the Berkeley Function-Calling Leaderboard.
  arXiv  Detail & Related papers  (2024-09-02T03:19:56Z)
• Fine-Tuning or Fine-Failing? Debunking Performance Myths in Large Language Models [0.8399688944263842]
  Large Language Models (LLMs) have the capability to understand and generate human-like text from input queries. This study extends this concept to the integration of LLMs within Retrieval-Augmented Generation (RAG) pipelines. We evaluate the impact of fine-tuning on the LLMs' capacity for data extraction and contextual understanding.
  arXiv  Detail & Related papers  (2024-06-17T04:35:17Z)
• Federated Full-Parameter Tuning of Billion-Sized Language Models with Communication Cost under 18 Kilobytes [53.4856038354195]
  Pre-trained large language models (LLMs) need fine-tuning to improve their responsiveness to natural language instructions. FedKSeed employs zeroth-order optimization with a finite set of random seeds. It significantly reduces transmission requirements between the server and clients to just a few random seeds.
  arXiv  Detail & Related papers  (2023-12-11T13:03:21Z)
• CMW-Net: Learning a Class-Aware Sample Weighting Mapping for Robust Deep Learning [55.733193075728096]
  Modern deep neural networks can easily overfit to biased training data containing corrupted labels or class imbalance. Sample re-weighting methods are popularly used to alleviate this data bias issue. We propose a meta-model capable of adaptively learning an explicit weighting scheme directly from data.
  arXiv  Detail & Related papers  (2022-02-11T13:49:51Z)
• Identifying Untrustworthy Samples: Data Filtering for Open-domain Dialogues with Bayesian Optimization [28.22184410167622]
  We present a data filtering method for open-domain dialogues. We score training samples with a quality measure, sort them in descending order, and filter out those at the bottom. Experimental results on two datasets show that our method can effectively identify untrustworthy samples.
  arXiv  Detail & Related papers  (2021-09-14T06:42:54Z)
• Bridging the Gap Between Clean Data Training and Real-World Inference for Spoken Language Understanding [76.89426311082927]
  Existing models are trained on clean data, which causes a textitgap between clean data training and real-world inference. We propose a method from the perspective of domain adaptation, by which both high- and low-quality samples are embedding into similar vector space. Experiments on the widely-used dataset, Snips, and large scale in-house dataset (10 million training examples) demonstrate that this method not only outperforms the baseline models on real-world (noisy) corpus but also enhances the robustness, that is, it produces high-quality results under a noisy environment.
  arXiv  Detail & Related papers  (2021-04-13T17:54:33Z)

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.