analiza aiqbench
SenseShift introduces an encoder-based framework that improves fine-grained sentiment control in text generation by utilizing bidirectional attention and iterative mask infilling, addressing limitations of standard decoder architectures. This approach enables precise sentence-level sentiment manipulation within long-form content while maintaining robustness against out-of-domain data. The research highlights a shift toward more controllable generative models capable of handling complex user constraints without sacrificing fluency.
SENSESHIFT: Continuous Sentiment-Controlled Text Generation via Encoder-based Mask Infilling
arXiv:2608.24304v1 Announce Type: cross Abstract: Recent controllable text generation (CTG) for sentiment control has largely focused on decoder-based large language models, making causal attention the dominant paradigm. While effective for fluent generation, these models still struggle to satisfy complex constraints and follow fine-grained sentiment signals specified by users. Existing sentiment-aware CTG methods typically simplify the problem by treating sentiment either as a coarse categorical label (e.g., positive or negative) or as a single fine-grained control signal applied to an entire document. Consequently, more challenging settings such as sentence-level sentiment control within long-form text remain underexplored. To address these limitations, we introduce SenseShift , an encoder-based framework for fine-grained sentence-level CTG. Unlike standard decoder architectures, SenseShift leverages bidirectional attention, quantized sentiment signals, and iterative mask infilling to generate local sentences conditioned on target sentiment intensity. Empirical evaluations on story and review generation demonstrate that SenseShift achieves stronger sentiment controllability while maintaining text quality and robustness to out-of-domain generation compared to larger decoder-based baselines.
Abstract:Recent controllable text generation (CTG) for sentiment control has largely focused on decoder-based large language models, making causal attention the dominant paradigm. While effective for fluent generation, these models still struggle to satisfy complex constraints and follow fine-grained sentiment signals specified by users. Existing sentiment-aware CTG methods typically simplify the problem by treating sentiment either as a coarse categorical label (e.g., positive or negative) or as a single fine-grained control signal applied to an entire document. Consequently, more challenging settings such as sentence-level sentiment control within long-form text remain underexplored. To address these limitations, we introduce SenseShift , an encoder-based framework for fine-grained sentence-level CTG. Unlike standard decoder architectures, SenseShift leverages bidirectional attention, quantized sentiment signals, and iterative mask infilling to generate local sentences conditioned on target sentiment intensity. Empirical evaluations on story and review generation demonstrate that SenseShift achieves stronger sentiment controllability while maintaining text quality and robustness to out-of-domain generation compared to larger decoder-based baselines.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2608.24304 [cs.CL]
(or arXiv:2608.24304v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.24304 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 255 cuvinte
Subiecte conexe
[AINews] High Taste Testers Ep. 1
Latent Space · 17 oct. 2026
[AINews] not much happened today
Latent Space · 3 oct. 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · 25 sept. 2026
Adaptive Parallel Reasoning: The Next Paradigm in Efficient Inference Scaling
BAIR Blog · 8 mai 2026