Score Before You Speak: Improving Persona Consistency in Dialogue Generation using Response Quality Scores
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912530150981632 |
|---|---|
| author | Saggar, Arpita Darling, Jonathan C. Dimitrova, Vania Sarikaya, Duygu Hogg, David C. |
| author_facet | Saggar, Arpita Darling, Jonathan C. Dimitrova, Vania Sarikaya, Duygu Hogg, David C. |
| contents | Persona-based dialogue generation is an important milestone towards building conversational artificial intelligence. Despite the ever-improving capabilities of large language models (LLMs), effectively integrating persona fidelity in conversations remains challenging due to the limited diversity in existing dialogue data. We propose a novel framework SBS (Score-Before-Speaking), which outperforms previous methods and yields improvements for both million and billion-parameter models. Unlike previous methods, SBS unifies the learning of responses and their relative quality into a single step. The key innovation is to train a dialogue model to correlate augmented responses with a quality score during training and then leverage this knowledge at inference. We use noun-based substitution for augmentation and semantic similarity-based scores as a proxy for response quality. Through extensive experiments with benchmark datasets (PERSONA-CHAT and ConvAI2), we show that score-conditioned training allows existing models to better capture a spectrum of persona-consistent dialogues. Our ablation studies also demonstrate that including scores in the input prompt during training is superior to conventional training setups. Code and further details are available at https://arpita2512.github.io/score_before_you_speak |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_06886 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Score Before You Speak: Improving Persona Consistency in Dialogue Generation using Response Quality Scores Saggar, Arpita Darling, Jonathan C. Dimitrova, Vania Sarikaya, Duygu Hogg, David C. Computation and Language Persona-based dialogue generation is an important milestone towards building conversational artificial intelligence. Despite the ever-improving capabilities of large language models (LLMs), effectively integrating persona fidelity in conversations remains challenging due to the limited diversity in existing dialogue data. We propose a novel framework SBS (Score-Before-Speaking), which outperforms previous methods and yields improvements for both million and billion-parameter models. Unlike previous methods, SBS unifies the learning of responses and their relative quality into a single step. The key innovation is to train a dialogue model to correlate augmented responses with a quality score during training and then leverage this knowledge at inference. We use noun-based substitution for augmentation and semantic similarity-based scores as a proxy for response quality. Through extensive experiments with benchmark datasets (PERSONA-CHAT and ConvAI2), we show that score-conditioned training allows existing models to better capture a spectrum of persona-consistent dialogues. Our ablation studies also demonstrate that including scores in the input prompt during training is superior to conventional training setups. Code and further details are available at https://arpita2512.github.io/score_before_you_speak |
| title | Score Before You Speak: Improving Persona Consistency in Dialogue Generation using Response Quality Scores |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2508.06886 |