Chengxiang Zhuo

Shuhao Chen, Weisen Jiang, Yeqi Gong et al.

Fine-tuning large language models often undermines their safety alignment, a problem further amplified by harmful fine-tuning attacks in which adversarial data removes safeguards and induces unsafe behaviors. We propose SPARD, a defense framework that integrates Safety-Projected Alternating optimization with Relevance-Diversity aware data selection. SPARD employs SPAG, which optimizes alternatively between utility updates and explicit safety projections with a set of safe data to enforce safety constraints. To curate safe data, we introduce a Relevance-Diversity Determinantal Point Process to select compact safe data, balancing task relevance and safety coverage. Experiments on GSM8K and OpenBookQA under four harmful fine-tuning attacks demonstrate that SPARD consistently achieves the lowest average attack success rates, substantially outperforming state-of-the-art defense methods, while maintaining high task accuracy. Code is available at https://github.com/shuhao02/SPARD.

Xinzhou Jin, Jintang Li, Liang Chen et al.

Graph neural networks (GNNs) have recently emerged as an effective approach to model neighborhood signals in collaborative filtering. Towards this research line, graph contrastive learning (GCL) demonstrates robust capabilities to address the supervision label shortage issue through generating massive self-supervised signals. Despite its effectiveness, GCL for recommendation suffers seriously from two main challenges: i) GCL relies on graph augmentation to generate semantically different views for contrasting, which could potentially disrupt key information and introduce unwanted noise; ii) current works for GCL primarily focus on contrasting representations using sophisticated networks architecture (usually deep) to capture high-order interactions, which leads to increased computational complexity and suboptimal training efficiency. To this end, we propose L2CL, a principled Layer-to-Layer Contrastive Learning framework that contrasts representations from different layers. By aligning the semantic similarities between different layers, L2CL enables the learning of complex structural relationships and gets rid of the noise perturbation in stochastic data augmentation. Surprisingly, we find that L2CL, using only one-hop contrastive learning paradigm, is able to capture intrinsic semantic structures and improve the quality of node representation, leading to a simple yet effective architecture. We also provide theoretical guarantees for L2CL in minimizing task-irrelevant information. Extensive experiments on five real-world datasets demonstrate the superiority of our model over various state-of-the-art collaborative filtering methods. Our code is available at https://github.com/downeykking/L2CL.

Yuanzhen Xie, Xinzhou Jin, Tao Xie et al.

In-context learning of large-language models (LLMs) has achieved remarkable success in the field of natural language processing, while extensive case studies reveal that the single-step chain-of-thought prompting approach faces challenges such as attention diffusion and inadequate performance in complex tasks like text-to-SQL. To improve the contextual learning capabilities of LLMs in text-to-SQL, a workflow paradigm method is proposed, aiming to enhance the attention and problem-solving scope of LLMs through decomposition. Specifically, the information determination module for eliminating redundant information and the brand-new prompt structure based on problem classification greatly enhance the model's attention. Additionally, the inclusion of self-correction and active learning modules greatly expands the problem-solving scope of LLMs, hence improving the upper limit of LLM-based approaches. Extensive experiments conducted on three datasets demonstrate that our approach outperforms other methods by a significant margin. About 2-3 percentage point improvements compared to the existing baseline on the Spider Dev, Spider-Realistic, and Bird Dev datasets and new SOTA results on the Spider Test dataset are achieved. Our code is available on GitHub: \url{https://github.com/FlyingFeather/DEA-SQL}.

Zhen Tao, Riwei Lai, Chenyun Yu et al.

Large language model (LLM) based recommendation agents personalize what they know through evolving per-user semantic memory, yet how they reason remains a universal, static system prompt shared identically across all users. This asymmetry is a fundamental bottleneck: when a recommendation fails, the agent updates its memory of user preferences but never interrogates the decision logic that produced the failure, leaving its reasoning process structurally unchanged regardless of how many mistakes it accumulates. To address this bottleneck, we propose SAGER (Self-Evolving Agent for Personalized Recommendation), the first recommendation agent framework in which each user is equipped with a dedicated policy skill, a structured natural-language document encoding personalized decision principles that evolves continuously through interaction. SAGER introduces a two-representation skill architecture that decouples a rich evolution substrate from a minimal inference-time injection, an incremental contrastive chain-of-thought engine that diagnoses reasoning flaws by contrasting accepted against unchosen items while preserving accumulated priors, and skill-augmented listwise reasoning that creates fine-grained decision boundaries where the evolved skill provides genuine discriminative value. Experiments on four public benchmarks demonstrate that SAGER achieves state-of-the-art performance, with gains orthogonal to memory accumulation, confirming that personalizing the reasoning process itself is a qualitatively distinct source of recommendation improvement.

Hongyang Su, Beibei Kong, Lei Cheng et al.

Conversational news recommendation requires grounding each suggestion in a rapidly evolving article corpus while addressing implicit user intents that lack explicit retrievable keywords. To characterize this scenario, we identify 6 intent types from production dialogues: five are implicit and pose fundamental challenges to standard RAG pipelines, forming a critical retrieve-first bottleneck. To address these issues, we introduce intent-driven Semantic ID (SID) generation under a Generate-then-Match paradigm. With two-stage training that consists of multi-task SID alignment and GPT-4 Chain-of-Thought distillation, an LLM maps diverse intents to hierarchical SID prefixes, which are then fuzzy-matched to the current news pool to guarantee fully grounded recommendations. Profile-Aware Dual-Signal Reasoning (PADR) further enables cold-start users to obtain valid recommendations using only profiles. On a mainstream Chinese news platform, our 7B model achieves 0% hallucination and 12.4% L1 match in the 152K open-generation SID space (4x random baseline). It matches GPT-4+Hybrid RAG on L1 while surpassing it on finer-grained metrics (L2 2x, Category +1.2pp) at ~100x lower cost. Cold-start users, where existing baselines score 0%, achieve 18.0% L1 (6x random), the highest among all user groups.

Yunsheng Pang, Zijian Liu, Yudong Li et al.

Slate recommendation, which presents users with a ranked item list in a single display, is ubiquitous across mainstream online platforms. Recent advances in generative models have shown significant potential for this task via autoregressive modeling of discrete semantic ID sequences. However, existing methods suffer from three key limitations: entangled item tokenization, inefficient sequential decoding, and the absence of holistic slate planning. These issues often result in substantial inference overhead and inadequate alignment with diverse user preferences and practical business requirements, hindering the industrial deployment of generative slate recommendation systems. In this paper, we propose HiGR, an efficient generative slate recommendation framework that integrates hierarchical planning with listwise preference alignment. First, we design an auto-encoder incorporating residual quantization and contrastive constraints, which tokenizes items into semantically structured IDs to enable controllable generation. Second, HiGR decouples the generation process into two stages: a list-level planning stage to capture global slate intent, and an item-level decoding stage to select specific items, effectively reducing the search space and enabling efficient generation. Third, we introduce a multi-objective and listwise preference alignment mechanism that enhances slate quality by leveraging implicit user feedback. Extensive experiments have validated the effectiveness of our HiGR method. Notably, it outperforms state-of-the-art baselines by over 10\% in offline recommendation quality while achieving a $5\times$ inference speedup. Furthermore, we have deployed HiGR on a commercial platform under Tencent (serving hundreds of millions of users), and online A/B tests show that it increases average watch time and average video plays by 1.22\% and 1.73\%, respectively.

Yuanzhen Xie, Tao Xie, Mingxiong Lin et al.

In most current research, large language models (LLMs) are able to perform reasoning tasks by generating chains of thought through the guidance of specific prompts. However, there still exists a significant discrepancy between their capability in solving complex reasoning problems and that of humans. At present, most approaches focus on chains of thought (COT) and tool use, without considering the adoption and application of human cognitive frameworks. It is well-known that when confronting complex reasoning challenges, humans typically employ various cognitive abilities, and necessitate interaction with all aspects of tools, knowledge, and the external environment information to accomplish intricate tasks. This paper introduces a novel intelligent framework, referred to as OlaGPT. OlaGPT carefully studied a cognitive architecture framework, and propose to simulate certain aspects of human cognition. The framework involves approximating different cognitive modules, including attention, memory, reasoning, learning, and corresponding scheduling and decision-making mechanisms. Inspired by the active learning mechanism of human beings, it proposes a learning unit to record previous mistakes and expert opinions, and dynamically refer to them to strengthen their ability to solve similar problems. The paper also outlines common effective reasoning frameworks for human problem-solving and designs Chain-of-Thought (COT) templates accordingly. A comprehensive decision-making mechanism is also proposed to maximize model accuracy. The efficacy of OlaGPT has been stringently evaluated on multiple reasoning datasets, and the experimental outcomes reveal that OlaGPT surpasses state-of-the-art benchmarks, demonstrating its superior performance. Our implementation of OlaGPT is available on GitHub: \url{https://github.com/oladata-team/OlaGPT}.

Xiangjin Xie, Yuxin Chen, Ruipeng Wang et al.

Graph neural networks have been widely used in recent recommender systems, where negative sampling plays an important role. Existing negative sampling methods restrict the relationship between nodes as either hard positive pairs or hard negative pairs. This leads to the loss of structural information, and lacks the mechanism to generate positive pairs for nodes with few neighbors. To overcome limitations, we propose a novel soft link-based sampling method, namely MixDec Sampling, which consists of Mixup Sampling module and Decay Sampling module. The Mixup Sampling augments node features by synthesizing new nodes and soft links, which provides sufficient number of samples for nodes with few neighbors. The Decay Sampling strengthens the digestion of graph structure information by generating soft links for node embedding learning. To the best of our knowledge, we are the first to model sampling relationships between nodes by soft links in GNN-based recommender systems. Extensive experiments demonstrate that the proposed MixDec Sampling can significantly and consistently improve the recommendation performance of several representative GNN-based models on various recommendation benchmarks.

Boyu Chen, Siran Chen, Zhengrong Yue et al.

User feedback is critical for refining recommendation systems, yet explicit feedback (e.g., likes or dislikes) remains scarce in practice. As a more feasible alternative, inferring user preferences from massive implicit feedback has shown great potential (e.g., a user quickly skipping a recommended video usually indicates disinterest). Unfortunately, implicit feedback is often noisy: a user might skip a video due to accidental clicks or other reasons, rather than disliking it. Such noise can easily misjudge user interests, thereby undermining recommendation performance. To address this issue, we propose a novel Group-aware User Behavior Simulation (G-UBS) paradigm, which leverages contextual guidance from relevant user groups, enabling robust and in-depth interpretation of implicit feedback for individual users. Specifically, G-UBS operates via two key agents. First, the User Group Manager (UGM) effectively clusters users to generate group profiles utilizing a ``summarize-cluster-reflect" workflow based on LLMs. Second, the User Feedback Modeler (UFM) employs an innovative group-aware reinforcement learning approach, where each user is guided by the associated group profiles during the reinforcement learning process, allowing UFM to robustly and deeply examine the reasons behind implicit feedback. To assess our G-UBS paradigm, we have constructed a Video Recommendation benchmark with Implicit Feedback (IF-VR). To the best of our knowledge, this is the first multi-modal benchmark for implicit feedback evaluation in video recommendation, encompassing 15k users, 25k videos, and 933k interaction records with implicit feedback. Extensive experiments on IF-VR demonstrate that G-UBS significantly outperforms mainstream LLMs and MLLMs, with a 4.0% higher proportion of videos achieving a play rate > 30% and 14.9% higher reasoning accuracy on IF-VR.

Shaojie Zhu, Zhaobin Wang, Chengxiang Zhuo et al.

CoT (Chain-of-Thought) is a way to solve reasoning problems for LLMs . Recently, many researches appear for improving the CoT capability of LLMs. In this work, we also proposed Olapa-MCoT, which is a LLMs based on llama2-13B PLM for finetuning and alignment learning. During the alignment training, we proposed the SimRRHF algorithm and Incorrect Data Relearning and mainly focused on optimizing the Chinese mathematical reasoning ability of Olapa-MCoT. The experiment achieved significant results, with the accuracy of Chinese mathematical reasoning up to 50%, 36% rise compared to llama2-13B. In addition, the accuracy of English reasoning ability also increased by nearly 4%.

Zhenpeng Li, Jianan Jiang, Yuhong Guo et al.

Domain adaptation, as a task of reducing the annotation cost in a target domain by exploiting the existing labeled data in an auxiliary source domain, has received a lot of attention in the research community. However, the standard domain adaptation has assumed perfectly observed data in both domains, while in real world applications the existence of missing data can be prevalent. In this paper, we tackle a more challenging domain adaptation scenario where one has an incomplete target domain with partially observed data. We propose an Incomplete Data Imputation based Adversarial Network (IDIAN) model to address this new domain adaptation challenge. In the proposed model, we design a data imputation module to fill the missing feature values based on the partial observations in the target domain, while aligning the two domains via deep adversarial adaption. We conduct experiments on both cross-domain benchmark tasks and a real world adaptation task with imperfect target domains. The experimental results demonstrate the effectiveness of the proposed method.