1
太阳光首次用于生成纠缠光子对，3 天户外实验保真度约 94%
量子纠缠光源新突破，太阳光户外生成光子对，保真度94%，登上《自然》！
IT之家 8 月 13 日消息，科技媒体 TechSpot 昨日（8 月 12 日）发布博文，报道称研究人员利用阳光生成纠缠光子对， 实验在户外连续 3 天完成，结果达到约 94% 保真度。 纠缠保真度 是衡量量子系统在传输、存储或运算过程中，多大程度上保持了理想 量子纠缠 特性的核心物理指标。 加…
2
Accurate Ensembles, Fragile Narratives: Multi-Scale Stacking and a Fidelity Audit of LLM-Generated Explanations for Credit Risk
从多尺度堆叠与保真度审计切入，揭示LLM解释信用风险评估模型的可靠性边界，值得AI+金融从业者关注。
arXiv:2608.08126v1 Announce Type: new Abstract: Credit scoring increasingly relies on models whose decision logic cannot be read off their parameters,…
3
我国科研团队破解超导量子计算“速度与保真度”矛盾
IT之家 8 月 7 日消息，据上海证券报今日报道，本源量子与中国科学技术大学联合研究团队创新性提出的“参数空间扩展受控相位门（PSE-CZ）”方案， 成功破解了超导量子计算领域长期存在的速度与保真度相互制约难题 ，让两比特量子门在保持高速的同时大幅提升精度。 相关成果已发表于国际物理顶级期刊《物理…
4
Out-Of-The-Loop Multi-Fidelity Bayesian Optimization
跳出循环的隐式监督，多保真贝叶斯优化迎来新范式，降本增效直击采样痛点。
arXiv:2608.04113v1 Announce Type: cross Abstract: Black-box optimization is a ubiquitous problem in science and engineering, often dealing with expens…
5
纠缠保真度超过 94%，量子互联网首次走出实验室
量子纠缠保真度超94%！这项突破首次在商用光纤上实现，为实用化量子互联网铺平道路。
IT之家 7 月 23 日消息，科技媒体 Phys 昨日（7 月 22 日）发布博文，报道称在一条承载着繁忙互联网流量的 24.4 公里商用光纤中， 美国西北大学科学家成功传输了纠缠光子对，同时保持了超过 94% 的纠缠保真度。 相关研究成果已于 7 月 20 日在期刊《Optica Quantum…
6
Will Scaling Improve Social Simulation with LLMs?
探讨LLM缩放能否提升社会模拟的保真度，揭示当前范式的局限与挑战
arXiv:2607.02464v1 Announce Type: new Abstract: Large Language Model (LLM) social simulations are a promising research method, but they are not yet fa…
7
Timesynth: A Temporal Fidelity Framework for Health Signal Digital Twins
论文提出Timesynth框架，专为健康信号数字孪生设计，解决时间保真度问题，推动精准医疗建模。
arXiv:2607.00431v1 Announce Type: new Abstract: Forecasting models for health-signal digital twins must preserve the oscillatory, frequency, phase, an…
8
Simulated Customers Never Walk Away: Decision Fidelity of LLM User Simulators Measured Against Real Purchase Outcomes
用真实购买结果衡量LLM用户模拟器的决策保真度，为对话AI评估提供新标尺
arXiv:2606.20708v1 Announce Type: new Abstract: LLM-as-user-simulation has become core infrastructure for conversational AI: agent benchmarks (tau-ben…
9
Displacement Is Not Direction: Evaluating Fidelity Metrics for Quantized LLM Deployment
量化模型部署的保真度评估，揭示位移≠方向的关键误区，助你选对指标。
arXiv:2606.19558v1 Announce Type: new Abstract: Fidelity metrics, such as per-token KL divergence (KLD) against a high-precision reference, are often …
10
A Mechanistic Understanding of Pronoun Fidelity in LLMs
深入剖析LLM代词保真失效的机制，揭示推理、重复与偏见间的复杂关系。
arXiv:2606.16407v1 Announce Type: cross Abstract: Faithful and robust pronoun use is important for fair and coherent generations, yet large language m…
11
Moment-Video: Diagnosing Temporal Fidelity of Video MLLMs on Momentary Visual Events
最新研究为视频多模态大模型的时间感知能力设计诊断基准，聚焦瞬时视觉事件，暴露模型在极短事件上的理解缺陷。
arXiv:2606.02522v1 Announce Type: cross Abstract: Video multimodal large language models (MLLMs) have made rapid progress on general and long-form vid…
12
Probing Embodied LLMs: When Higher Observation Fidelity Hurts Problem Solving
提升观察精度反而降低问题解决能力——这项研究挑战了具身LLM的传统认知，揭示保真度与推理之间的意外权衡。
arXiv:2605.20072v1 Announce Type: new Abstract: Large Language Models are increasingly proposed as cognitive components for robotic systems, yet their…
13
Fidelity Probes for Specification--Code Alignment
论文提出规范性代码对齐的保真度探测方法，为新模型验证提供新思路。
arXiv:2605.17246v1 Announce Type: new Abstract: We introduce fidelity probes: natural-language questions generated from a reference artifact with code…
14
Dimension-Level Intent Fidelity Evaluation for Large Language Models: Evidence from Structured Prompt Ablation
提出维度级意图保真度评估框架，通过结构化提示消融实验揭示LLM的意图还原与形式复制差异。
arXiv:2605.14517v1 Announce Type: cross Abstract: Holistic evaluation scores capture overall output quality but do not distinguish whether a model rep…