The correct model response in the face of uncertainty is to abstain from answering a question so as not to mislead the user. In this work, we study the ability of LLMs to abstain from answering context-dependent science questions when provided insufficient or incorrect context. We probe model sensitivity in several settings: removing gold context, replacing gold context with irrelevant context, and providing additional context beyond what is given. In experiments on four QA datasets with four LLMs, we show that performance varies greatly across models, across the type of context provided, and also by question type; in particular, many LLMs seem unable to abstain from answering boolean questions using standard QA prompts. Our analysis also highlights the unexpected impact of abstention performance on QA task accuracy. Counter-intuitively, in some settings, replacing gold context with irrelevant context or adding irrelevant context to gold context can improve abstention performance in a way that results in improvements in task performance. Our results imply that changes are needed in QA dataset design and evaluation to more effectively assess the correctness and downstream impacts of model abstention.
In this paper, we build a visual dialogue dataset, named InfoVisDial, which provides rich informative answers in each round even with external knowledge related to the visual content. Different from existing datasets where the answer is compact and short, InfoVisDial contains long free-form answers with rich information in each round of dialogue. For effective data collection, the key idea is to bridge the large-scale multimodal model (e.g., GIT) and the language models (e.g., GPT-3). GIT can describe the image content even with scene text, while GPT-3 can generate informative dialogue based on the image description and appropriate prompting techniques. With such automatic pipeline, we can readily generate informative visual dialogue data at scale. Then, we ask human annotators to rate the generated dialogues to filter the low-quality conversations.Human analyses show that InfoVisDial covers informative and diverse dialogue topics: $54.4\%$ of the dialogue rounds are related to image scene texts, and $36.7\%$ require external knowledge. Each round's answer is also long and open-ended: $87.3\%$ of answers are unique with an average length of $8.9$, compared with $27.37\%$ and $2.9$ in VisDial. Last, we propose a strong baseline by adapting the GIT model for the visual dialogue task and fine-tune the model on InfoVisDial. Hopefully, our work can motivate more effort on this direction.
Our paper aims to generate diverse and realistic animal motion sequences from textual descriptions, without a large-scale animal text-motion dataset. While the task of text-driven human motion synthesis is already extensively studied and benchmarked, it remains challenging to transfer this success to other skeleton structures with limited data. In this work, we design a model architecture that imitates Generative Pretraining Transformer (GPT), utilizing prior knowledge learned from human data to the animal domain. We jointly train motion autoencoders for both animal and human motions and at the same time optimize through the similarity scores among human motion encoding, animal motion encoding, and text CLIP embedding. Presenting the first solution to this problem, we are able to generate animal motions with high diversity and fidelity, quantitatively and qualitatively outperforming the results of training human motion generation baselines on animal data. Additionally, we introduce AnimalML3D, the first text-animal motion dataset with 1240 animation sequences spanning 36 different animal identities. We hope this dataset would mediate the data scarcity problem in text-driven animal motion generation, providing a new playground for the research community.
Growing attention has been paid in Conversational Recommendation System (CRS), which works as a conversation-based and recommendation task-oriented tool to provide items of interest and explore user preference. However, existing work in CRS fails to explicitly show the reasoning logic to users and the whole CRS still remains a black box. Therefore we propose a novel end-to-end framework named Explanation Generation for Conversational Recommendation (EGCR) based on generating explanations for conversational agents to explain why they make the action. EGCR incorporates user reviews to enhance the item representation and increase the informativeness of the whole conversation. To the best of our knowledge, this is the first framework for explainable conversational recommendation on real-world datasets. Moreover, we evaluate EGCR on one benchmark conversational recommendation datasets and achieve better performance on both recommendation accuracy and conversation quality than other state-of-the art models. Finally, extensive experiments demonstrate that generated explanations are not only having high quality and explainability, but also making CRS more trustworthy. We will make our code available to contribute to the CRS community
As recommender systems become increasingly sophisticated and complex, they often suffer from lack of fairness and transparency. Providing robust and unbiased explanations for recommendations has been drawing more and more attention as it can help address these issues and improve trustworthiness and informativeness of recommender systems. However, despite the fact that such explanations are generated for humans who respond more strongly to messages with appropriate emotions, there is a lack of consideration for emotions when generating explanations for recommendations. Current explanation generation models are found to exaggerate certain emotions without accurately capturing the underlying tone or the meaning. In this paper, we propose a novel method based on a multi-head transformer, called Emotion-aware Transformer for Explainable Recommendation (EmoTER), to generate more robust, fair, and emotion-enhanced explanations. To measure the linguistic quality and emotion fairness of the generated explanations, we adopt both automatic text metrics and human perceptions for evaluation. Experiments on three widely-used benchmark datasets with multiple evaluation metrics demonstrate that EmoTER consistently outperforms the existing state-of-the-art explanation generation models in terms of text quality, explainability, and consideration for fairness to emotion distribution. Implementation of EmoTER will be released as an open-source toolkit to support further research.