arXiv subjects

Chenke Yin

Publications and source records attributed to Chenke Yin.

3 recordsLinked to original sources

arXiv2025

A Hierarchical Quantized Tokenization Framework for Task-Adaptive Graph Representation Learning

Foundation models in language and vision benefit from a unified discrete token interface that converts raw inputs into sequences for scalable pre-training and inference. For graphs, an effective tokenizer should yield reusable discrete codes that capture both node semantics and relational structure across scales, yet prior quantization-based graph tokenizers typically combine residual vector quantization (RVQ) levels with fixed rules and often focus on a single structural view, limiting cross-task transfer. We present a hierarchical quantized tokenization framework with task-conditioned routing and dual-view token streams. It produces multi-scale codes and two synchronized sequences: a local stream that preserves node-level information and a diffusion-style multi-hop stream that summarizes connectivity. A lightweight router learns task-dependent mixtures over RVQ depths to select an appropriate granularity, while a gated cross-attention module aligns and fuses the two streams into a single token sequence without altering the downstream backbone encoder. Experiments on node classification and link prediction show consistent gains over strong quantized baselines at matched compute, with ablations verifying contributions from hierarchical quantization, adaptive routing, and fusion.

cs.IR↗

arXiv2025

Uncertainty-Aware Semantic Decoding for LLM-Based Sequential Recommendation

Large language models have been widely applied to sequential recommendation tasks, yet during inference, they continue to rely on decoding strategies developed for natural language processing. This creates a mismatch between text-generation objectives and recommendation next item selection objectives. This paper addresses this limitation by proposing an Uncertainty-aware Semantic Decoding (USD) framework that combines logit-based clustering with adaptive scoring to improve next-item predictions. Our approach clusters items with similar logit vectors into semantic equivalence groups, then redistributes probability mass within these clusters and computes entropy across them to control item scoring and sampling temperature during recommendation inference. Experiments on Amazon Product datasets (six domains) gains of 18.5\% in HR@3, 11.9\% in NDCG@3, and 10.8\% in MRR@3 compared to state-of-the-art baselines. Hyperparameter analysis confirms the optimal parameters among various settings, and experiments on H\&M, and Netflix datasets indicate that the framework can adapt to differing recommendation domains. The experimental results confirm that integrating semantic clustering and uncertainty assessment yields more reliable and accurate recommendations.

cs.IR↗

arXiv2025

Harnessing Light for Cold-Start Recommendations: Leveraging Epistemic Uncertainty to Enhance Performance in User-Item Interactions

Most recent paradigms of generative model-based recommendation still face challenges related to the cold-start problem. Existing models addressing cold item recommendations mainly focus on acquiring more knowledge to enrich embeddings or model inputs. However, many models do not assess the efficiency with which they utilize the available training knowledge, leading to the extraction of significant knowledge that is not fully used, thus limiting improvements in cold-start performance. To address this, we introduce the concept of epistemic uncertainty to indirectly define how efficiently a model uses the training knowledge. Since epistemic uncertainty represents the reducible part of the total uncertainty, we can optimize the recommendation model further based on epistemic uncertainty to improve its performance. To this end, we propose a Cold-Start Recommendation based on Epistemic Uncertainty (CREU) framework. Additionally, CREU is inspired by Pairwise-Distance Estimators (PaiDEs) to efficiently and accurately measure epistemic uncertainty by evaluating the mutual information between model outputs and weights in high-dimensional spaces. The proposed method is evaluated through extensive offline experiments on public datasets, which further demonstrate the advantages and robustness of CREU. The source code is available at https://github.com/EsiksonX/CREU.

cs.IR↗