Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment

Alignment in artificial intelligence pursues the consistency between model responses and human preferences as well as values. In practice, the multifaceted nature of human preferences inadvertently introduces what is known as the "alignment tax" -a compromise where enhancements in alignment within one objective (e.g.,harmlessness) can diminish performance in others (e.g.,helpfulness). However, existing alignment techniques are mostly unidirectional, leading to suboptimal trade-offs and poor flexibility over various objectives. To navigate this challenge, we argue the prominence of grounding LLMs with evident preferences. We introduce controllable preference optimization (CPO), which explicitly specifies preference scores for different objectives, thereby guiding the model to generate responses that meet the requirements. Our experimental analysis reveals that the aligned models can provide responses that match various preferences among the "3H" (helpfulness, honesty, harmlessness) desiderata. Furthermore, by introducing diverse data and alignment goals, we surpass baseline methods in aligning with single objectives, hence mitigating the impact of the alignment tax and achieving Pareto improvements in multi-objective alignment.

FreeFlow: A Comprehensive Understanding on Diffusion Probabilistic Models via Optimal Transport

The blooming diffusion probabilistic models (DPMs) have garnered significant interest due to their impressive performance and the elegant inspiration they draw from physics. While earlier DPMs relied upon the Markovian assumption, recent methods based on differential equations have been rapidly applied to enhance the efficiency and capabilities of these models. However, a theoretical interpretation encapsulating these diverse algorithms is insufficient yet pressingly required to guide further development of DPMs. In response to this need, we present FreeFlow, a framework that provides a thorough explanation of the diffusion formula as time-dependent optimal transport, where the evolutionary pattern of probability density is given by the gradient flows of a functional defined in Wasserstein space. Crucially, our framework necessitates a unified description that not only clarifies the subtle mechanism of DPMs but also indicates the roots of some defects through creative involvement of Lagrangian and Eulerian views to understand the evolution of probability flow. We particularly demonstrate that the core equation of FreeFlow condenses all stochastic and deterministic DPMs into a single case, showcasing the expansibility of our method. Furthermore, the Riemannian geometry employed in our work has the potential to bridge broader subjects in mathematics, which enable the involvement of more profound tools for the establishment of more outstanding and generalized models in the future.

DiFace: Cross-Modal Face Recognition through Controlled Diffusion

Diffusion probabilistic models (DPMs) have exhibited exceptional proficiency in generating visual media of outstanding quality and realism. Nonetheless, their potential in non-generative domains, such as face recognition, has yet to be thoroughly investigated. Meanwhile, despite the extensive development of multi-modal face recognition methods, their emphasis has predominantly centered on visual modalities. In this context, face recognition through textual description presents a unique and promising solution that not only transcends the limitations from application scenarios but also expands the potential for research in the field of cross-modal face recognition. It is regrettable that this avenue remains unexplored and underutilized, a consequence from the challenges mainly associated with three aspects: 1) the intrinsic imprecision of verbal descriptions; 2) the significant gaps between texts and images; and 3) the immense hurdle posed by insufficient databases.To tackle this problem, we present DiFace, a solution that effectively achieves face recognition via text through a controllable diffusion process, by establishing its theoretical connection with probability transport. Our approach not only unleashes the potential of DPMs across a broader spectrum of tasks but also achieves, to the best of our knowledge, a significant accuracy in text-to-image face recognition for the first time, as demonstrated by our experiments on verification and identification.

Generative Flow Network for Listwise Recommendation

Personalized recommender systems fulfill the daily demands of customers and boost online businesses. The goal is to learn a policy that can generate a list of items that matches the user's demand or interest. While most existing methods learn a pointwise scoring model that predicts the ranking score of each individual item, recent research shows that the listwise approach can further improve the recommendation quality by modeling the intra-list correlations of items that are exposed together. This has motivated the recent list reranking and generative recommendation approaches that optimize the overall utility of the entire list. However, it is challenging to explore the combinatorial space of list actions and existing methods that use cross-entropy loss may suffer from low diversity issues. In this work, we aim to learn a policy that can generate sufficiently diverse item lists for users while maintaining high recommendation quality. The proposed solution, GFN4Rec, is a generative method that takes the insight of the flow network to ensure the alignment between list generation probability and its reward. The key advantages of our solution are the log scale reward matching loss that intrinsically improves the generation diversity and the autoregressive item selection model that captures the item mutual influences while capturing future reward of the list. As validation of our method's effectiveness and its superior diversity during active exploration, we conduct experiments on simulated online environments as well as an offline evaluation framework for two real-world datasets.

Exploration and Regularization of the Latent Action Space in Recommendation

In recommender systems, reinforcement learning solutions have effectively boosted recommendation performance because of their ability to capture long-term user-system interaction. However, the action space of the recommendation policy is a list of items, which could be extremely large with a dynamic candidate item pool. To overcome this challenge, we propose a hyper-actor and critic learning framework where the policy decomposes the item list generation process into a hyper-action inference step and an effect-action selection step. The first step maps the given state space into a vectorized hyper-action space, and the second step selects the item list based on the hyper-action. In order to regulate the discrepancy between the two action spaces, we design an alignment module along with a kernel mapping function for items to ensure inference accuracy and include a supervision module to stabilize the learning process. We build simulated environments on public datasets and empirically show that our framework is superior in recommendation compared to standard RL baselines.

APAN: Asynchronous Propagation Attention Network for Real-time Temporal Graph Embedding

Limited by the time complexity of querying k-hop neighbors in a graph database, most graph algorithms cannot be deployed online and execute millisecond-level inference. This problem dramatically limits the potential of applying graph algorithms in certain areas, such as financial fraud detection. Therefore, we propose Asynchronous Propagation Attention Network, an asynchronous continuous time dynamic graph algorithm for real-time temporal graph embedding. Traditional graph models usually execute two serial operations: first graph computation and then model inference. We decouple model inference and graph computation step so that the heavy graph query operations will not damage the speed of model inference. Extensive experiments demonstrate that the proposed method can achieve competitive performance and 8.7 times inference speed improvement in the meantime.

APAN: Asynchronous Propagate Attention Network for Real-time Temporal Graph Embedding

Limited by the time complexity of querying k-hop neighbors in a graph database, most graph algorithms cannot be deployed online and execute millisecond-level inference. This problem dramatically limits the potential of applying graph algorithms in certain areas, such as financial fraud detection. Therefore, we propose Asynchronous Propagate Attention Network, an asynchronous continuous time dynamic graph algorithm for real-time temporal graph embedding. Traditional graph models usually execute two serial operations: first graph computation and then model inference. We decouple model inference and graph computation step so that the heavy graph query operations will not damage the speed of model inference. Extensive experiments demonstrate that the proposed method can achieve competitive performance and 8.7 times inference speed improvement in the meantime.