The paper introduces the concept of epistemic warrant to assess when users should rely on individual recommendations from large language models (LLMs). It defines a four-tier reliance certificate—unstable, context-dependent, locally supported, and broadly supported—based on the stability and scope of a model’s preference. Experiments show that these warrants align with expert judgments and crowd consensus, and they provide distinct information from confidence scores or decision difficulty.
By Shai Vardi, Jo\~ao Sedoc
The paper investigates how large language models (LLMs) can evaluate explanations in recommender systems. It generates 18 explanation prototypes and has 14 LLMs rate them, comparing the results to human ratings from a user study. Findings show that while LLMs mimic human rating patterns and correlate moderately with human judgments, their absolute agreement is low and varies with model size and evaluation design, leading to four practical recommendations for using LLMs in this context.
By Kathrin Wardatzky, Oana Inel, Luca Rossetto, Abraham Bernstein
RPCBench is a new benchmark designed to evaluate large language models’ ability to critique recommendation requests by detecting, diagnosing, and handling flawed premises. It includes evidence‑grounded test instances across five recommendation domains and ten types of premise failures, and introduces a fine‑grained evaluation framework covering detection, error localization, handling strategy, and evidence faithfulness. Experiments with 11 LLMs reveal that proactive detection is the main bottleneck, with models struggling most on underspecified‑premise errors and showing that optimal critique quality occurs at intermediate reasoning lengths.
By Zhongru Chen, Yuan Wu, Yi Chang
arXiv:2607. 20529v1 Announce Type: cross Abstract: Large Language Model (LLM) ensembles are increasingly used to improve reliability by combining predictions from multiple LLMs.
By Jiawei Zheng, Jiazhen Zhang
ReFIne is a training framework that augments large reasoning models with three trustworthiness properties: interpretability, faithfulness, and reliability. It combines supervised fine‑tuning with GRPO to produce structured, tag‑based reasoning traces, explicitly disclose decisive information, and provide self‑assessments of soundness and confidence. Applied to Qwen3 models, ReFIne improves interpretability by 44.0 %, faithfulness by 18.8 %, and reliability by 42.4 % on mathematical benchmarks.
By Chung-En Sun, Ge Yan, Akshay Kulkarni, Tsui-Wei Weng
The paper introduces PROVE-REC, a two‑pass framework that generates verifiable preference proofs for large language model (LLM) recommendation systems. In Pass A, the model condenses user interaction histories into a compact proof of positive and avoidance claims linked to specific evidence entries. Pass B then uses only this proof and its evidence to predict the next item, ensuring the recommendation follows the reasoning path. Verification steps compare masked evidence and removed claims to confirm grounding and influence, while a ranking‑preservation objective retains useful historical information. Experiments on diverse real‑world datasets show PROVE‑REC outperforms strong baselines by up to 7.45%, producing claims that are both better grounded and more influential to recommendation quality.
By Yu Hou, Nathaniel Kang, Pengkai Wang, Hua Li
EviGraph is a system that enables proof‑carrying selective recommendation over temporal public‑service knowledge graphs. It distinguishes between critical decision requirements and information that can remain unresolved, linking the former to evidence via a language agent and verifying support with a deterministic checker. Experiments on a bilingual Hong Kong public‑service benchmark show that this approach reduces unnecessary abstention, while additional verification can actually withdraw supported recommendations without improving decision quality.
By Yixi Zhou, Sikun Wang, Lei Fan, Fan Zhang
arXiv:2608. 06377v1 Announce Type: cross Abstract: Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong.
By Xian Sun, Wei Chow, Yingshuo Wang, Junhao Liu, Wei Gao, Qing Wu, Lingdong Kong
The paper presents a method for fine‑tuning a large language model (LLM) recommender to generate personalized, non‑harmful explanations for its recommendations. By training two LLM‑judge reward models and using constrained GRPO, the authors achieve a significant increase in the PASS rate for all three criteria, from 0.649 to 0.956 on their own judges and from 0.677 to 0.931 on an independent judge. The fine‑tuned model maintains its original recommendation performance, demonstrating that LLM‑based recommenders can be adapted to complex tasks without loss of effectiveness.
By Jiashu He, Emma Yanyang Kong, JJ Tan, David Fagnan
The paper introduces a Calibrated Reflection approach to improve confidence estimation in Large Language Models (LLMs). It combines structured reasoning with a distance‑aware calibration technique, featuring a Maximum Confidence Selection method, a reflection‑based prompting mechanism, and an ordinal‑aware calibration strategy. Experiments on datasets such as HelpSteer2, Llama T‑REx, and a proprietary conversational set show the method works for both conversational and fact‑based classification tasks.
By Umesh Bodhwani, Yuan Ling, Shujing Dong, Yarong Feng, Hongfei Li, Ayush Goyal
arXiv:2610.29294v1 Announce Type: cross
Abstract: Large language models (LLMs) are increasingly used to support consequential decisions, yet their reliability under equivalent representations of unce...
By Yuyao Zhang, Yakun Cui, Han Zhu, Jinsong Lu, Yike Guo, Sirui Han
The paper introduces NPRec, a model‑agnostic framework that uses counterfactual reasoning to neutralize popularity bias in large language model–based recommender systems. By generating debiased textual guidelines that separate intrinsic user interests from popularity signals, NPRec injects these guidelines at inference time to guide the LLM’s generation without updating parameters. Experiments on three real‑world datasets show improved recommendation accuracy, explanation quality, and debiasing performance.
By Guanrong Li, Haolin Yang, Xinyu Liu, Zhen Wu, Rui Xia, Xinyu Dai