paper-with-me

홈 › Papers

Semantic Space Grounded Weighted Decoding for Multi-Attribute Controllable Dialogue Generation

2023-05-04 · Zhiling Zhang, Mengyue Wu, Kenny Q. Zhu

Controlling chatbot utterance generation with multiple attributes such as personalities, emotions and dialogue acts is a practically useful but under-studied problem. We propose a novel framework called DASC that possesses strong controllability with a weighted decoding paradigm, while improving generation quality with the grounding in an attribute semantics space. Generation with multiple attributes is then intuitively implemented with an interpolation of multiple attribute embeddings, which results in substantial reduction in the model sizes. Experiments show that DASC can achieve high control accuracy in generation task with the simultaneous control of 3 aspects while also producing interesting and reasonably sensible responses, even in an out-of-distribution robustness test.

📄 PDF Abstract BibTeX arXiv:2305.02820

Code (1)

blmoistawinde/dasc 공식 구현 pytorch

Tasks

AttributeChatbotDialogue Generation

Similar Papers 제목 키워드 기반

Breaking the Illusion: When Positive Meets Negative in Multimodal Decoding

2026-04-22 · Yubo Jiang, Yitong An, Xin Yang, Abudukelimu Wuerkaixi 외 arxiv

Vision-Language Models (VLMs) are frequently undermined by object hallucination, generating content that contradicts visual reality, due to an over-reliance on linguistic priors. We introduce Positive-and-Negative Decodi…

The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information Steering

2025-02-05 · Zhuowei Li, Haizhou Shi, Yunhe Gao, Di Liu 외

Large Vision-Language Models (LVLMs) can reason effectively over both textual and visual inputs, but they tend to hallucinate syntactically coherent yet visually ungrounded contents. In this paper, we investigate the int…

Hallucination

Revisit What You See: Disclose Language Prior in Vision Tokens for Efficient Guided Decoding of LVLMs

2025-06-11 · Beomsik Cho, Jaehyung Kim

Large Vision-Language Models (LVLMs) have demonstrated remarkable performance across various multimodal tasks by integrating visual perception with language understanding. However, conventional decoding strategies of LVL…

HallucinationObject HallucinationText GenerationVisual Grounding

Sequence-Level Certainty Reduces Hallucination In Knowledge-Grounded Dialogue Generation

2023-10-28 · Yixin Wan, Fanyou Wu, Weijie Xu, Srinivasan H. Sengamedu

In this work, we propose sequence-level certainty as a common theme over hallucination in Knowledge Grounded Dialogue Generation (KGDG). We explore the correlation between the level of hallucination in model responses an…

Dialogue GenerationHallucinationText Generation

DOPRA: Decoding Over-accumulation Penalization and Re-allocation in Specific Weighting Layer

2024-07-21 · Jinfeng Wei, Xiaofeng Zhang

In this work, we introduce DOPRA, a novel approach designed to mitigate hallucinations in multi-modal large language models (MLLMs). Unlike existing solutions that typically involve costly supplementary training data or …