MOSS (Model Sanity Service) is a production framework that embeds LLM-based relevance scoring into a continuous monitoring loop for recommendation, search, and offer surfaces, using targeted sampling, strategy-aware prompting, and structured alerting to diagnose semantic failures. It has been deployed for over six months at a large U.S. retailer, identifying two confirmed serving defects and quantifying a 76.4% reduction in alert rate for a backfill blending fix. Benchmark results show that LLM judges outperform non-LLM baselines on both recommendation and search tasks, demonstrating that a single LLM monitoring framework can deliver actionable quality signals across multiple product discovery surfaces.
By Alina Nesen, Prathyusha Kanmanth Reddy, David Relyea, Rohit Avinash Mujumdar, Samaneh Karimi, Bo Shen, Jordan Hashemi
The paper examines LLM-based recommendation rerankers that are often evaluated under an oracle protocol, which guarantees the ground-truth item is present in the scored set. Across Amazon datasets, this protocol overestimates realistic NDCG@10 by 92–95% because realistic retrieval only covers 2–19% of relevant items at K=100, creating a recall ceiling that limits any closed-candidate reranker's top‑k NDCG. The authors find that various optimisation strategies—including prompt engineering, model scaling, sequential models, supervised neural rerankers, LoRA fine‑tuning, hybrid retrieval, score‑aware prompting, and LLM+CF fusion—do not significantly improve over a collaborative‑filtering baseline under realistic retrieval, and they propose a Recall‑Aware Evaluation Protocol (RAEP) to better assess rerankers in low‑recall regimes.
By Zhaohui Wang
arXiv:2608. 15780v1 Announce Type: cross Abstract: Stale recommendations are a pervasive challenge and a leading source of user complaints on large-scale content platforms.
By Di Bai, Feng Han, Zhenwei Tang, Jintao Liu, Luoshu Wang, Jialu Liu
The paper introduces a retrieval‑grounded credit‑assignment method for generative recommenders that use Semantic IDs (SIDs). By structuring each autoregressive trace into a history summary, a set of interest hypotheses, and a final SID, a frozen retriever verifies each hypothesis as a catalog query. Rewards are assigned at the hypothesis level when any query retrieves the target within the top‑K, allowing distinct updates for rollouts that share the same SID reward and improving SID recommendation performance on Amazon Reviews datasets.
arXiv:2606. 29947v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as rerankers in recommender systems, with the expectation that semantic understanding will help in cold-start and long-tail regimes.
By Zhe Dong (University of Maine at Presque Isle), Fang Qin (Stanford University), Manish Shah (Independent Researcher), Yicheng Wang (Independent Researcher)
The paper investigates the trade‑off of using a shared search‑and‑recommendation index that scores new items purely from features, thereby keeping the index open to unseen items. Experiments on public logs show that a feature‑based tower can match warm‑item performance (Recall@20 0.9595 vs 0.9510) and a lexical baseline, while a full‑catalog check is inconclusive. The study also quantifies the cost of this openness on recommendation quality across several baselines, revealing that exact full‑softmax training improves recall but is impractical at catalog scale.
By Theodore Rogers, Joe Standerfer, Dmitrii Timoshenko, Haoxue Li, Zuhaib Akhtar, Soyoung Yang