Zhengyi Yang

Junkang Wu, Yuexiang Xie, Zhengyi Yang et al.

Direct Preference Optimization (DPO) has emerged as a compelling approach for training Large Language Models (LLMs) to adhere to human preferences. However, the performance of DPO is sensitive to the fine-tuning of its trade-off parameter $β$, as well as to the quality of the preference data. We analyze the impact of $β$ and data quality on DPO, uncovering that optimal $β$ values vary with the informativeness of pairwise data. Addressing the limitations of static $β$ values, we introduce a novel framework that dynamically calibrates $β$ at the batch level, informed by data quality considerations. Additionally, our method incorporates $β$-guided data filtering to safeguard against the influence of outliers. Through empirical evaluation, we demonstrate that our dynamic $β$ adjustment technique significantly improves DPO's performance across a range of models and datasets, offering a more robust and adaptable training paradigm for aligning LLMs with human feedback. The code is available at \url{https://github.com/junkangwu/beta-DPO}.

Junkang Wu, Yuexiang Xie, Zhengyi Yang et al.

This study addresses the challenge of noise in training datasets for Direct Preference Optimization (DPO), a method for aligning Large Language Models (LLMs) with human preferences. We categorize noise into pointwise noise, which includes low-quality data points, and pairwise noise, which encompasses erroneous data pair associations that affect preference rankings. Utilizing Distributionally Robust Optimization (DRO), we enhance DPO's resilience to these types of noise. Our theoretical insights reveal that DPO inherently embeds DRO principles, conferring robustness to pointwise noise, with the regularization coefficient $β$ playing a critical role in its noise resistance. Extending this framework, we introduce Distributionally Robustifying DPO (Dr. DPO), which integrates pairwise robustness by optimizing against worst-case pairwise scenarios. The novel hyperparameter $β'$ in Dr. DPO allows for fine-tuned control over data pair reliability, providing a strategic balance between exploration and exploitation in noisy training environments. Empirical evaluations demonstrate that Dr. DPO substantially improves the quality of generated text and response accuracy in preference datasets, showcasing enhanced performance in both noisy and noise-free settings. The code is available at https://github.com/junkangwu/Dr_DPO.

Yiwei Yan, Hao Li, Hua He et al.

Online medical consultations generate large volumes of conversational health data that often embed protected health information, requiring robust methods to classify data categories and assign risk levels in line with policies and practice. However, existing approaches lack unified standards and reliable automated methods to fulfill sensitivity classification for such conversational health data. This study presents a large language model-based extraction pipeline, SALP-CG, for classifying and grading privacy risks in online conversational health data. We concluded health-data classification and grading rules in accordance with GB/T 39725-2020. Combining few-shot guidance, JSON Schema constrained decoding, and deterministic high-risk rules, the backend-agnostic extraction pipeline achieves strong category compliance and reliable sensitivity across diverse LLMs. On the MedDialog-CN benchmark, models yields robust entity counts, high schema compliance, and accurate sensitivity grading, while the strongest model attains micro-F1=0.900 for maximum-level prediction. The category landscape stratified by sensitivity shows that Level 2-3 items dominate, enabling re-identification when combined; Level 4-5 items are less frequent but carry outsize harm. SALP-CG reliably helps classify categories and grading sensitivity in online conversational health data across LLMs, offering a practical method for health data governance. Code is available at https://github.com/dommii1218/SALP-CG.

Chengpeng Li, Zhengyi Yang, Jizhi Zhang et al.

Reinforcement learning (RL) has been widely applied in recommendation systems due to its potential in optimizing the long-term engagement of users. From the perspective of RL, recommendation can be formulated as a Markov decision process (MDP), where recommendation system (agent) can interact with users (environment) and acquire feedback (reward signals).However, it is impractical to conduct online interactions with the concern on user experience and implementation complexity, and we can only train RL recommenders with offline datasets containing limited reward signals and state transitions. Therefore, the data sparsity issue of reward signals and state transitions is very severe, while it has long been overlooked by existing RL recommenders.Worse still, RL methods learn through the trial-and-error mode, but negative feedback cannot be obtained in implicit feedback recommendation tasks, which aggravates the overestimation problem of offline RL recommender. To address these challenges, we propose a novel RL recommender named model-enhanced contrastive reinforcement learning (MCRL). On the one hand, we learn a value function to estimate the long-term engagement of users, together with a conservative value learning mechanism to alleviate the overestimation problem.On the other hand, we construct some positive and negative state-action pairs to model the reward function and state transition function with contrastive learning to exploit the internal structure information of MDP. Experiments demonstrate that the proposed method significantly outperforms existing offline RL and self-supervised RL methods with different representative backbone networks on two real-world datasets.

Junkang Wu, Xue Wang, Zhengyi Yang et al.

Aligning large language models (LLMs) with human values and intentions is crucial for their utility, honesty, and safety. Reinforcement learning from human feedback (RLHF) is a popular approach to achieve this alignment, but it faces challenges in computational efficiency and training stability. Recent methods like Direct Preference Optimization (DPO) and Simple Preference Optimization (SimPO) have proposed offline alternatives to RLHF, simplifying the process by reparameterizing the reward function. However, DPO depends on a potentially suboptimal reference model, and SimPO's assumption of a fixed target reward margin may lead to suboptimal decisions in diverse data settings. In this work, we propose $α$-DPO, an adaptive preference optimization algorithm designed to address these limitations by introducing a dynamic reward margin. Specifically, $α$-DPO employs an adaptive preference distribution, balancing the policy model and the reference model to achieve personalized reward margins. We provide theoretical guarantees for $α$-DPO, demonstrating its effectiveness as a surrogate optimization objective and its ability to balance alignment and diversity through KL divergence control. Empirical evaluations on AlpacaEval 2 and Arena-Hard show that $α$-DPO consistently outperforms DPO and SimPO across various model settings, establishing it as a robust approach for fine-tuning LLMs. Our method achieves significant improvements in win rates, highlighting its potential as a powerful tool for LLM alignment. The code is available at https://github.com/junkangwu/alpha-DPO

Yutian Liu, Zhengyi Yang, Jiancan Wu et al.

Recent advances have applied large language models (LLMs) to sequential recommendation, leveraging their pre-training knowledge and reasoning capabilities to provide more personalized user experiences. However, existing LLM-based methods fail to sufficiently leverage the rich temporal information inherent in users' historical interaction sequences, stemming from fundamental architectural constraints: LLMs process information through self-attention mechanisms that lack inherent sequence ordering and rely on position embeddings designed primarily for natural language rather than user interaction sequences. This limitation significantly impairs their ability to capture the evolution of user preferences over time and predict future interests accurately. To address this critical gap, we propose \underline{C}ounterfactual \underline{E}nhanced \underline{T}emporal Framework for LLM-Based \underline{Rec}ommendation (CETRec). CETRec is grounded in causal inference principles, which allow it to isolate and measure the specific impact of temporal information on recommendation outcomes. Combined with our counterfactual tuning task derived from causal analysis, CETRec effectively enhances LLMs' awareness of both absolute order (how recently items were interacted with) and relative order (the sequential relationships between items). Extensive experiments on real-world datasets demonstrate the effectiveness of our CETRec. Our code is available at https://anonymous.4open.science/r/CETRec-B9CE/.

Yuxin Chen, Junfei Tan, An Zhang et al.

Recommender systems aim to predict personalized rankings based on user preference data. With the rise of Language Models (LMs), LM-based recommenders have been widely explored due to their extensive world knowledge and powerful reasoning abilities. Most of the LM-based recommenders convert historical interactions into language prompts, pairing with a positive item as the target response and fine-tuning LM with a language modeling loss. However, the current objective fails to fully leverage preference data and is not optimized for personalized ranking tasks, which hinders the performance of LM-based recommenders. Inspired by the current advancement of Direct Preference Optimization (DPO) in human preference alignment and the success of softmax loss in recommendations, we propose Softmax-DPO (S-DPO) to instill ranking information into the LM to help LM-based recommenders distinguish preferred items from negatives, rather than solely focusing on positives. Specifically, we incorporate multiple negatives in user preference data and devise an alternative version of DPO loss tailored for LM-based recommenders, which is extended from the traditional full-ranking Plackett-Luce (PL) model to partial rankings and connected to softmax sampling strategies. Theoretically, we bridge S-DPO with the softmax loss over negative sampling and find that it has an inherent benefit of mining hard negatives, which assures its exceptional capabilities in recommendation tasks. Empirically, extensive experiments conducted on three real-world datasets demonstrate the superiority of S-DPO to effectively model user preference and further boost recommendation performance while providing better rewards for preferred items. Our codes are available at https://github.com/chenyuxin1999/S-DPO.