Industrial generative recommendation systems operating over discrete Semantic IDs (SIDs) are largely behavior-driven, and thus struggle to proactively activate latent demand before explicit user signals emerge, leading to intent cold-start. To address this, we propose RIGER (Reasoning-grounded Intent injection for GE nerative Recommendation), a deployable two-stage framework that integrates offline large language model (LLM) reasoning into an online generative recommender under strict latency constraints. Offline, to ensure scalable deployment, we distill the latent-intent inference capability of a strong LLM into a lightweight forecasting model using an automated data curation pipeline---leveraging judge-guided prompt calibration and future-query-guided rejection filtering. Online, to bridge the representation mismatch between free-form textual intents and the discrete SID token space, predicted intents are converted into SID-native tokens through a behavior-grounded mapping and injected into the deployed decoder-only retrieval backbone. We further fine-tune the model with beam-aware GRPO, introducing a hierarchical intent-alignment exploration reward in SID space while preserving exploitation behavior through KL regularization. Offline evaluations demonstrate a substantial increase in intent-aligned density and diversity with only a marginal reduction in hindsight recall, indicating that RIGER effectively enhances proactive intent exploration while preserving its capability to exploit historical behaviors. In a large-scale e-commerce display advertising system, RIGER improves clicks by 1.6% and advertiser spend by 1.3%.
Xusong Chen, Peini Guo, Fang Liu et al.· Annual International ACM SIG...· 0 citations
Chimeric antigen receptor (CAR)-T cell therapy has transformed the treatment landscape for relapsed or refractory hematologic malignancies, producing high remission rates in otherwise treatment-resistant patients. However, primary resistance and disease relapse remain common, particularly in solid tumors, limiting long-term benefit and broader clinical applicability. As the population of patients failing therapy grows, there is an urgent need for an integrated understanding of resistance mechanisms and a structured approach to salvage therapy. This review proposes a conceptual “Why-How-What if” framework to navigate the complexities of treatment failure. We first address “Why” therapy fails, identifying multifactorial drivers including tumor-intrinsic factors like antigen loss and immune evasion, T cell-intrinsic dysfunction such as exhaustion and limited persistence, and extrinsic constraints imposed by an immunosuppressive tumor microenvironment. We then explore “How” to enhance efficacy through mechanism-based strategies. These include rational combination approaches with immune checkpoint inhibitors or small molecule inhibitors, and next-generation engineering such as dual-target, armored, and in vivo generated CAR-T cells aimed at overcoming metabolic and physical barriers. Finally, we address the “What if” of treatment failure by summarizing individualized salvage options, for which current clinical evidence is derived predominantly from hematologic malignancies. These strategies range from target-switching and bispecific antibodies to emerging cellular platforms like CAR-natural killer cells and consolidation via allogeneic hematopoietic stem cell transplantation. By integrating mechanisms of failure with evolving optimization and salvage strategies, this framework provides a practical roadmap for clinical and translational progress. Future success will depend on biomarker-guided combinations and the continued diversification of adoptive cell therapy platforms.
Bi-Jing Wu, Jia-Hui Wang, Qihua Zou et al.· Biomarker Research· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
