CoFree framework addresses reasoning collapse in LLM embedding learning, achieving a 2.8 nDCG@10 improvement.

Zihan Gong, Xiaohan Ye, Jiangchao Yao et al.

MERIT-Rank enhances text reranking through multi-perspective reasoning space and progressive optimization, with the 4B model outperforming 7B and 32B models on BRIGHT.

Lijun Liu, Zhengzong Chen, Wenyan Li et al.

MSS-Complement method recovers state-conditioned minimal sufficient evidence for coding agents, achieving 73.0% coverage at five items.

Zhexi Feng, Ruiyi Zhang, Yongbo Yang et al.

Using natural-language user profiles, UPR method improves transparency in recommendations, validated on Amazon and TripAdvisor datasets.

Noah Mamié, Laurin van den Bergh

Dense feature representation and optimization boost CVR prediction AUC to 0.828535.

Yi Zhang, Weiliang Ji

SELF-INDEX framework autonomously evolves search indices, improving retrieval performance on datasets like BRIGHT.

Sangam Lee, Wonjae Lee, Sunghwan Kim et al.

G3RAG uses geometric gain graphs for zero-token construction, enhancing multi-hop RAG with a 4.26 F1 score increase.

Zeliang Li, Xiaofen Xing, Kailing Guo et al.

Embedding Surgery: Adaptive ranking correction in dense retrieval, achieving 60.64% nDCG@10 improvement.

Maddalena Amendola, Antonio Mallia, Raffaele Perego

AlleCompanion framework combines Two Tower architecture and ComCat mapping for improved recommendation accuracy.

Aleksandra Osowska-Kurczab, Klaudia Nazarko, Eliška Kosturová et al.

Repeated queries exhaust an LLM's brand recommendations but not its sources.

Dmitrij Żatuchin

SAM-D2Q enhances e-commerce search with multimodal document expansion, boosting GMV by 3.38%.

Hui Zhou, Jian Hui Ji, Lei Ma et al.

The Dice Roll Method offers a standardized protocol for auditing LLM brand recommendations, accurately predicting reliability.

Dmitrij Żatuchin

UniCon employs a hierarchical, context-centric architecture, improving CTR prediction by 0.0139 AUC and online metrics by over 3%.

Jiajun Cui, Zhengqi Xu, Fan Zhang et al.

OrthoRec enhances multimodal recommendation via orthogonal purification and topology-guided MoE, excelling on Amazon datasets.

Jialin Liu, Zhaorui Zhang, Ray C. C. Cheung

PersonaGen-1M uses MinHash LSH and semantic deduplication to build 1.03M intent-annotated buyer personas.

Dmitrij Żatuchin, Daniil Dzemesjuk

RePair method improves cross-modal retrieval accuracy on Flickr30K and COCO30K by repairing retrieval errors.

Siyi Liu, Xiaorong Zhu, Enjun Du et al.

QUEST combines synthetic query generation, topic extraction, and LLM-based relevance assessment to identify credibility indicators in asylum appeal documents, achieving MAP@100 of 0.572 and 0.589 on datasets.

Maria Vlachou, Anna Murphy Høgenhaug, Mohammad N. S. Jahromi et al.

Combines LLMs and textual embeddings for long-form article-video matching, boosting user engagement.

Arnaud Corone, Brice Pierre de la Briere, Gladys Roch et al.

HubMixer introduces learnable latent hubs for efficient feature interaction, reducing parameters by 50% while outperforming SOTA models.

Jie Zhou, Zixian Gong, Wenhao Li et al.

AMUR employs an information-theoretic approach for selective modality-interest alignment, achieving 5-8% NDCG improvements in multimodal recommendation.

Wenze Ma, Chenyu Sun, Yanmin Zhu et al.