Palette of Language Models: A Solver for Controlled Text Generation

• URL: http://arxiv.org/abs/2503.11182v1
• Date: Fri, 14 Mar 2025 08:30:09 GMT
• Title: Palette of Language Models: A Solver for Controlled Text Generation
• Authors: Zhe Yang, Yi Huang, Yaqin Chen, Xiaoting Wu, Junlan Feng, Chao Deng,
• Abstract summary: Large language models can produce controlled texts that closely adhere to specific requirements when prompted appropriately.<n>A common approach is to linearly combine single-attribute models, but this strategy often overlooks attribute overlaps and can lead to conflicts.<n>We propose a novel combination strategy inspired by the Law of Total Probability and Conditional Mutual Information Minimization on generative language models.
• Score: 20.774257685046994
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advancements in large language models have revolutionized text generation with their remarkable capabilities. These models can produce controlled texts that closely adhere to specific requirements when prompted appropriately. However, designing an optimal prompt to control multiple attributes simultaneously can be challenging. A common approach is to linearly combine single-attribute models, but this strategy often overlooks attribute overlaps and can lead to conflicts. Therefore, we propose a novel combination strategy inspired by the Law of Total Probability and Conditional Mutual Information Minimization on generative language models. This method has been adapted for single-attribute control scenario and is termed the Palette of Language Models due to its theoretical linkage between attribute strength and generation style, akin to blending colors on an artist's palette. Moreover, positive correlation and attribute enhancement are advanced as theoretical properties to guide a rational combination strategy design. We conduct experiments on both single control and multiple control settings, and achieve surpassing results.

Related papers

• Towards Lightweight, Adaptive and Attribute-Aware Multi-Aspect Controllable Text Generation with Large Language Models [40.54453001537357]
  Multi-aspect controllable text generation aims to control text generation in attributes from multiple aspects.<n>Supervised fine-tuning methods are often employed for this task due to their simplicity and effectiveness.<n>We propose a lightweight, adaptive and attribute-aware framework for multi-aspect controllable text generation.
  arXiv  Detail & Related papers  (2025-02-19T06:56:02Z)
• Multi-Attribute Constraint Satisfaction via Language Model Rewriting [67.5778646504987]
  Multi-Attribute Constraint Satisfaction (MACS) is a method capable of finetuning language models to satisfy user-specified constraints on multiple external real-value attributes. Our work opens new avenues for generalized and real-value multi-attribute control, with implications for diverse applications spanning NLP and bioinformatics.
  arXiv  Detail & Related papers  (2024-12-26T12:36:39Z)
• P-Masking: Power Law Masking Improves Multi-attribute Controlled Generation [14.763505073094779]
  LingGen is a novel approach for controlled text generation that offers precise control over a wide array of linguistic attributes. LingGen employs a dynamic P-MASKING strategy, which samples masking rates from a power law distribution during training. Experiments demonstrate that LingGen surpasses current state-of-the-art models in both attribute control accuracy and text fluency.
  arXiv  Detail & Related papers  (2024-10-31T17:55:45Z)
• Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directions [20.351245266660378]
  Recent advances in text-to-image (T2I) diffusion models have significantly improved the quality of generated images.<n>Providing efficient control over individual subjects, particularly the attributes characterizing them, remains a key challenge.<n>No current approach offers both simultaneously, resulting in a gap when trying to achieve precise continuous and subject-specific attribute modulation.
  arXiv  Detail & Related papers  (2024-03-25T18:00:42Z)
• Toward Unified Controllable Text Generation via Regular Expression Instruction [56.68753672187368]
  Our paper introduces Regular Expression Instruction (REI), which utilizes an instruction-based mechanism to fully exploit regular expressions' advantages to uniformly model diverse constraints. Our method only requires fine-tuning on medium-scale language models or few-shot, in-context learning on large language models, and requires no further adjustment when applied to various constraint combinations.
  arXiv  Detail & Related papers  (2023-09-19T09:05:14Z)
• Click: Controllable Text Generation with Sequence Likelihood Contrastive Learning [69.35360098882606]
  We introduce Click for controllable text generation, which needs no modification to the model architecture. It employs a contrastive loss on sequence likelihood, which fundamentally decreases the generation probability of negative samples. It also adopts a novel likelihood ranking-based strategy to construct contrastive samples from model generations.
  arXiv  Detail & Related papers  (2023-06-06T01:56:44Z)
• Language Model Cascades [72.18809575261498]
  Repeated interactions at test-time with a single model, or the composition of multiple models together, further expands capabilities. Cases with control flow and dynamic structure require techniques from probabilistic programming. We formalize several existing techniques from this perspective, including scratchpads / chain of thought, verifiers, STaR, selection-inference, and tool use.
  arXiv  Detail & Related papers  (2022-07-21T07:35:18Z)
• Mix and Match: Learning-free Controllable Text Generation using Energy Language Models [33.97800741890231]
  We propose Mix and Match LM, a global score-based alternative for controllable text generation. We interpret the task of controllable generation as drawing samples from an energy-based model. We use a Metropolis-Hastings sampling scheme to sample from this energy-based model.
  arXiv  Detail & Related papers  (2022-03-24T18:52:09Z)
• Controllable Natural Language Generation with Contrastive Prefixes [120.12778570283956]
  GPT2 generation utilizes a set of small attribute-specific vectors, called prefixes, to steer natural language generation. We propose a novel supervised method and also an unsupervised method to train the prefixes for single-aspect control. Experimental results on both single-aspect and multi-aspect control show that our methods can guide generation towards the desired attributes while keeping high linguistic quality.
  arXiv  Detail & Related papers  (2022-02-27T00:31:03Z)
• Controllable Dialogue Generation with Disentangled Multi-grained Style Specification and Attribute Consistency Reward [47.96949534259019]
  We propose a controllable dialogue generation model to steer response generation under multi-attribute constraints. We categorize the commonly used control attributes into global and local ones, which possess different granularities of effects on response generation. Our model can significantly outperform competitive baselines in terms of response quality, content diversity and controllability.
  arXiv  Detail & Related papers  (2021-09-14T14:29:38Z)
• Attribute Alignment: Controlling Text Generation from Pre-trained Language Models [46.19190007510232]
  We propose a simple and flexible method for controlling text generation by aligning disentangled attribute representations. In contrast to recent efforts on training a discriminator to perturb the token level distribution for an attribute, we use the same data to learn an alignment function to guide the pre-trained, non-controlled language model to generate texts with the target attribute without changing the original language model parameters.
  arXiv  Detail & Related papers  (2021-03-20T01:51:32Z)

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.