aiqbench analysis
This research addresses the critical challenge of noise accumulation in user profiles within conversational AI systems by introducing a dense retrieval framework that jointly optimizes profile denoising and relevance modeling. By moving beyond sparse token selection, the proposed method leverages latent embedding spaces to more accurately identify beneficial personalization signals for complex information needs. The approach demonstrates superior performance on the iKAT 2026 benchmark, highlighting advancements in personalized retrieval architectures.
RALI@iKAT 2026: Retrieval-aware User Profile Denoising via Personalized Conversational Dense Retrieval
arXiv:2610.17673v1 Announce Type: cross Abstract: Modern conversational information access systems continuously extract user information from conversations to build user profiles for personalized retrieval. As user profiles accumulate over time, they inevitably become noisy, making effective user profile denoising a fundamental challenge. Existing personalized conversational query reformulation methods rely on the black-box reasoning capability of LLMs to expand the original query with relevant user profile phrases, producing a verbose query for retrieval. Such rewrite-then-retrieve pipelines reduce user profile denoising to the explicit selection of profile phrases in the sparse token space, separating personalization from downstream relevance modeling. Consequently, they often fail to identify user information that is truly beneficial for retrieval. To enable retrieval-aware user profile denoising, we propose Personalized Conversational Dense Retrieval (PCDR), which learns to jointly project the conversation history and user profile into a latent embedding space dedicated to retrieval. By inheriting the context denoising capability from conversational dense retrievers, PCDR requires only a handful of personalized relevance annotations for training. We evaluate PCDR on the iKAT 2026 tasks, where it serves as the core retrieval component of a conversational AI agent assisting users with complex information needs.
Abstract:Modern conversational information access systems continuously extract user information from conversations to build user profiles for personalized retrieval. As user profiles accumulate over time, they inevitably become noisy, making effective user profile denoising a fundamental challenge. Existing personalized conversational query reformulation methods rely on the black-box reasoning capability of LLMs to expand the original query with relevant user profile phrases, producing a verbose query for retrieval. Such rewrite-then-retrieve pipelines reduce user profile denoising to the explicit selection of profile phrases in the sparse token space, separating personalization from downstream relevance modeling. Consequently, they often fail to identify user information that is truly beneficial for retrieval. To enable retrieval-aware user profile denoising, we propose Personalized Conversational Dense Retrieval (PCDR), which learns to jointly project the conversation history and user profile into a latent embedding space dedicated to retrieval. By inheriting the context denoising capability from conversational dense retrievers, PCDR requires only a handful of personalized relevance annotations for training. We evaluate PCDR on the iKAT 2026 tasks, where it serves as the core retrieval component of a conversational AI agent assisting users with complex information needs.
12 pages, 4 figures. Presented at the SCAI Workshop at SIGIR 2026. Ranked 1st in both retrieval performance and response quality at iKAT 2026
Information Retrieval (cs.IR); Computation and Language (cs.CL)
arXiv:2610.17673 [cs.IR]
(or arXiv:2610.17673v1 [cs.IR] for this version)
https://doi.org/10.48550/arXiv.2610.17673 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 307 words
Related coverage
GEAK V3: Agent-Driven, Repository-Level GPU Kernel Optimization across HIP, Triton, and FlyDSL on AMD GPUs
AMD · Jul 20, 2026
TimeRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation
arXiv · Aug 25, 2026
GitLab 19.4 release notes
GitLab · Sep 17, 2026
SatIR: Scalable High-Recall Constraint-Satisfaction-Based Information Retrieval for Clinical Trials Matching
arXiv · Sep 16, 2026