Direct Preference Optimization (DPO) has gained significant attention for its simplicity and computational efficiency in aligning large language models (LLMs). Recent advancements have extended DPO to multimodal scenarios, achieving strong performance. However, traditional DPO relies on binary preference optimization, rewarding or penalizing entire responses without considering fine-grained segment correctness, leading to suboptimal solutions. The root of this issue lies in the absence of fine-grained supervision during the optimization process. To address this, we propose Adaptive Sentence-level Preference Optimization (ASPO), which evaluates individual sentences for more precise preference optimization. By dynamically calculating adaptive rewards at the sentence level based on model predictions, ASPO enhances response content assessment without additional models or parameters. This significantly improves the alignment of multimodal features. Extensive experiments show that ASPO substantially enhances the overall performance of multimodal models.

The growing demand for larger-scale models in the development of Large Language Models (LLMs) poses challenges for efficient training within limited computational resources. Traditional fine-tuning methods often exhibit instability in multi-task learning and rely heavily on extensive training resources. Here, we propose MoDULA (Mixture of Domain-Specific and Universal LoRA), a novel Parameter Efficient Fine-Tuning (PEFT) Mixture-of-Expert (MoE) paradigm for improved fine-tuning and parameter efficiency in multi-task learning. The paradigm effectively improves the multi-task capability of the model by training universal experts, domain-specific experts, and routers separately. MoDULA-Res is a new method within the MoDULA paradigm, which maintains the model’s general capability by connecting universal and task-specific experts through residual connections. The experimental results demonstrate that the overall performance of the MoDULA-Flan and MoDULA-Res methods surpasses that of existing fine-tuning methods on various LLMs. Notably, MoDULA-Res achieves more significant performance improvements in multiple tasks while reducing training costs by over 80% without losing general capability. Moreover, MoDULA displays flexible pluggability, allowing for the efficient addition of new tasks without retraining existing experts from scratch. This progressive training paradigm circumvents data balancing issues, enhancing training efficiency and model stability. Overall, MoDULA provides a scalable, cost-effective solution for fine-tuning LLMs with enhanced parameter efficiency and generalization capability.

The effectiveness of Large Language Models (LLMs) relies on their capacity to understand instructions and generate human-like responses. However, aligning LLMs with complex human preferences remains a significant challenge due to the potential misinterpretation of user prompts. Current methods for aligning LLM behaviors fall into two categories: output optimization (such as RLHF, RLAIF, and DPO) and input optimization (like OPRO and BPO). While both approaches aim to guide LLMs towards generating responses that align with desired objectives, the labor-intensive and intentions-inconsistent data annotation, as well as the strict and tedious training supervision, make them struggle to yield optimal results across all models. To address these shortcomings, we introduce a novel self-renewal approach called Prompt Optimization with Implicit Reasoning (POIR). It consists of two key components: 1) a model-specific and self-recirculating data collection method that leverages self-evaluation to enhance prompts in accordance with the model’s intrinsic logits, and 2) a prompt rewrite schema that injects implicit reasoning for direct preference learning. Through self-renewal optimization, POIR refines LLM outputs to better align with human preferences across various LLMs and tasks, without relying on supervised fine-tuning. Extensive experiments on a range of LLMs and tasks demonstrate POIR’s superior performance. We believe this advancement offers a novel paradigm for developing LLMs that are more attuned to user intentions.

Leveraging the dependency tree of the input sentence is able to improve the model performance for relation extraction. A challenging issue is how to remove confusions from the tree. Efforts have been made to utilize the dependency connections between words to selectively emphasize target-relevant information. However, these approaches are limited in focusing on exploiting dependency types. In this paper, we propose dependency position encoding (DPE), an efficient way of incorporating both dependency connections and dependency types into the self-attention mechanism to distinguish the importance of different word dependencies for the task. In contrast to previous studies that process input sentence and dependency information in separate streams, DPE can be seamlessly incorporated into the Transformer and makes it possible to use an one-stream scheme to extract relations between entity pairs. Extensive experiments show that models with our DPE significantly outperform the previous methods on SemEval 2010 Task 8, KBP37, and TACRED.

In e-commerce system, category prediction is to automatically predict categories of given texts. Different from traditional classification where there are no relations between classes, category prediction is reckoned as a standard hierarchical classification problem since categories are usually organized as a hierarchical tree. In this paper, we address hierarchical category prediction. We propose a Deep Hierarchical Classification framework, which incorporates the multi-scale hierarchical information in neural networks and introduces a representation sharing strategy according to the category tree. We also define a novel combined loss function to punish hierarchical prediction losses. The evaluation shows that the proposed approach outperforms existing approaches in accuracy.