Yang Yang

Toby Jia-Jun Li, Yuwen Lu, Jaylexia Clark et al.

The growing inequality in gig work between workers and platforms has become a critical social issue as gig work plays an increasingly prominent role in the future of work. The AI inequality is caused by (1) the technology divide in who has access to AI technologies in gig work; and (2) the data divide in who owns the data in gig work leads to unfair working conditions, growing pay gap, neglect of workers' diverse preferences, and workers' lack of trust in the platforms. In this position paper, we argue that a bottom-up approach that empowers individual workers to access AI-enabled work planning support and share data among a group of workers through a network of end-user-programmable intelligent assistants is a practical way to bridge AI inequality in gig work under the current paradigm of privately owned platforms. This position paper articulates a set of research challenges, potential approaches, and community engagement opportunities, seeking to start a dialogue on this important research topic in the interdisciplinary CHIWORK community.

Boyong Wu, Chao Yan, Chen Hu et al.

This paper presents Step-Audio 2, an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation. By integrating a latent audio encoder and reasoning-centric reinforcement learning (RL), Step-Audio 2 achieves promising performance in automatic speech recognition (ASR) and audio understanding. To facilitate genuine end-to-end speech conversation, Step-Audio 2 incorporates the generation of discrete audio tokens into language modeling, significantly enhancing its responsiveness to paralinguistic information such as speaking styles and emotions. To effectively leverage the rich textual and acoustic knowledge in real-world data, Step-Audio 2 integrates retrieval-augmented generation (RAG) and is able to call external tools such as web search to mitigate hallucination and audio search to switch timbres. Trained on millions of hours of speech and audio data, Step-Audio 2 delivers intelligence and expressiveness across diverse conversational scenarios. Evaluation results demonstrate that Step-Audio 2 achieves state-of-the-art performance on various audio understanding and conversational benchmarks compared to other open-source and commercial solutions. Please visit https://github.com/stepfun-ai/Step-Audio2 for more information.

Jia-Qi Yang, Chenglei Dai, Dan OU et al.

With the advancement of multimedia internet, the impact of visual characteristics on the decision of users to click or not within the online retail industry is increasingly significant. Thus, incorporating visual features is a promising direction for further performance improvements in click-through rate (CTR). However, experiments on our production system revealed that simply injecting the image embeddings trained with established pre-training methods only has marginal improvements. We believe that the main advantage of existing image feature pre-training methods lies in their effectiveness for cross-modal predictions. However, this differs significantly from the task of CTR prediction in recommendation systems. In recommendation systems, other modalities of information (such as text) can be directly used as features in downstream models. Even if the performance of cross-modal prediction tasks is excellent, it is challenging to provide significant information gain for the downstream models. We argue that a visual feature pre-training method tailored for recommendation is necessary for further improvements beyond existing modality features. To this end, we propose an effective user intention reconstruction module to mine visual features related to user interests from behavior histories, which constructs a many-to-one correspondence. We further propose a contrastive training method to learn the user intentions and prevent the collapse of embedding vectors. We conduct extensive experimental evaluations on public datasets and our production system to verify that our method can learn users' visual interests. Our method achieves $0.46\%$ improvement in offline AUC and $0.88\%$ improvement in Taobao GMV (Cross Merchandise Volume) with p-value$<$0.01.

Bowen Dong, Jiawei Zhang, Chenwei Zhang et al.

Online knowledge libraries refer to the online data warehouses that systematically organize and categorize the knowledge-based information about different kinds of concepts and entities. In the era of big data, the setup of online knowledge libraries is an extremely challenging and laborious task, in terms of efforts, time and expense required in the completion of knowledge entities. Especially nowadays, a large number of new knowledge entities, like movies, are keeping on being produced and coming out at a continuously accelerating speed, which renders the knowledge library setup and completion problem more difficult to resolve manually. In this paper, we will take the online movie knowledge libraries as an example, and study the "Multiple aligned ISomeric Online Knowledge LIbraries Completion problem" (Miso-Klic) problem across multiple online knowledge libraries. Miso-Klic aims at identifying the missing entities for multiple knowledge libraries synergistically and ranking them for editing based on certain ranking criteria. To solve the problem, a thorough investigation of two isomeric online knowledge libraries, Douban and IMDB, have been carried out in this paper. Based on analyses results, a novel deep online knowledge library completion framework "Integrated Deep alignEd Auto-encoder" (IDEA) is introduced to solve the problem. By projecting the entities from multiple isomeric knowledge libraries to a shared feature space, IDEA solves the Miso-Klic problem via three steps: (1) entity feature space unification via embedding, (2) knowledge library fusion based missing entity identification, and (3) missing entity ranking. Extensive experiments done on the real-world online knowledge library dataset have demonstrated the effectiveness of IDEA in addressing the problem.