🌳 任务树 🛠 任务管理 📄 全部 tiddler

#lesson

报价页外发版的反向审稿结论 20260425183347000
ABT-263→OSK 间隔的保守锚定原则 20260425183256000
报价页外发版的歧义消除顺序 20260425183118000
一页式报价页的最小骨架 20260425182651000
一页式成功费条款模板 20260425182525000
成功费触发条件必须同时约束结果变量与测量质量 20260425182313000
桥接 offer 的最小可验证结构 20260425182107000
结果导向/分成式报价的零EV门槛通常更苛刻 20260425181936000
零EV来源占比误差预算表 20260425181651000
顺序干预的洗脱间隔先用药代锚定保守下界 20260425181535000
来源归因误差预算 20260425181416000
皮肤中 senolytic→OSK 的最小实验蓝图 20260425181207000
来源占比误差会把零EV边界直接推翻 20260425181053000
固定范围报价的零EV门槛近似随价格反比缩放 20260425180847000
渠道零EV门槛表（保守重算） 20260425180829000
OSK vs senolytic 的最小可比实验规格 20260425180717000
部分重编程 go/no-go 判据与安全边界 20260425180607000
研究型服务的产品形态选择 20260425180506000
固定范围服务的零EV重算应先锁价格档再核验来源混合 20260425180339000
皮肤顺序联合的最小分组骨架 20260425180046000
来源分布的最小可核验采样口径 20260425175718000
senolytic→OSK 顺序联合的最小实验设计缺口 20260425175624000
senolytic→OSK 顺序联合证据缺口的三态判定语句 20260425175244000
来源分布对零EV边界高度敏感 20260425175033000
senolytic→OSK 顺序联合的原始证据边界 20260425174632000
自噬轴候选应分为“独立干预、上游状态与 readout”三层 20260425174343000
成年哺乳动物自噬增强候选的审计结论 20260425174141000
首个可售服务的报价/交付决策器 20260425174118000
自噬增强原始正例检索的噪音类型 20260425173833000
成年哺乳动物自噬增强正例的统一审计边界 20260425173655000
固定范围服务报价必须先过正EV门槛 20260425173625000
TFEB-mediated autophagy can extend lifespan in middle-aged mice 20260425173437000
20人样本一页式决策器 20260425173357000
20人样本下的分段门槛决策表 20260425173312000
小样本二项判定门槛 20260425173138000
小样本首轮高信号阈值的基准率稳健性 20260425173006000
原始 greedy witness span 外的双交换稳定性 20260425173000000
最小样本获客实验的二项判定门槛 20260425172827000
各渠道的零EV成交率门槛 20260425172711000
获客渠道的线索温度阶梯 20260425172527000
锚点不变量检验要先排除触锚编辑，再用最小 witness span 收缩候选域 20260425172439000
检索要用字段收缩来验证结构性空集 20260425172341000
序贯干预必须显式检索“先后顺序” 20260425172311000
可访问 API 优先于网页搜索作为灰色文献入口 20260425172041000
研究型服务三件套足以覆盖客户决策动作 20260425172007000
研究型服务的边界语言 20260425171934000
研究型服务交付物必须驱动客户决策 20260425171814000
骨相关系统中 senolytic→OSK 顺序实验的空检索判定 20260425171703000
研究型服务的最小证据包 20260425171611000
骨相关系统中的 senolytic→OSK 顺序实验仍未检出 20260425171453000
研究型服务的隐藏沟通/修订缓冲检验 20260425171317000
空检索也是证据结论 20260425171236000
非糖尿病急性骨折/肌腱修复中的 senolytic 证据缺口 20260425171209000
急性骨/肌腱修复中的 senolytic 方向依赖组织与背景 20260425171003000
公开锚点缺失时用成本下界先定价 20260425170854000
疾病背景必须与组织和终点一起匹配 20260425170821000
senolytic 在非糖尿病急性神经损伤中可能促再生而非单向延迟修复 20260425170752000
研究型服务的最小交付包 20260425170723000
交付证据、签收动作与交接清单的最小集合 20260425170617000
研究型服务的明确排除项 20260425170438000
senolytic 对急性修复并非必然有害 20260425170413000
重编程中的 senescence 可能是阶段依赖而非单向有害 20260425170201000
单一组织内 senescence 负担态与生态位态的可证伪判别框架 20260425165902000
研究型产品化服务的EV门槛 20260425165539000
跨组织 head-to-head 证据缺口的并行空检索 20260425165254000
warm lead 载体路径必须先做可访问性验证 20260425165203000
warm lead 原始记录的外部载体先于核验 20260425165018000
warm lead 原始记录必须先定位真实载体 20260425164825000
骨组织中 OSK vs senolytic 的比较证据缺失边界 20260425164526000
三层保守下界的联合EV判据 20260425164214000
senolytic→OSK 顺序实验的最小证据门槛 20260425164134000
小样本正EV判断应使用保守置信下界 20260425163950000
Autophagy is a baseline-positive mammalian longevity node 20260425163912000
warm lead 原始记录的最小可核验样本模板 20260425163845000
自噬轴第三条成年哺乳动物原始正例的审计边界 20260425163721000
warm lead 的最低回复阈值由资格率与成交率共同决定 20260425163654000
首批 warm lead 回复率不能单独证明正EV 20260425163044000
间接自噬关联证据不应计入直接自噬增强正例 20260425162314000
固定范围诊断型服务的报价应先过正EV检查 20260425162206000
升级路径与加购条件应以现金期望值和高信号为门槛 20260425161839000
试单优惠必须以首单现金期望值为约束 20260425161506000
单一原始正例不足以升级为普适候选 20260425161427000
最小交付边界必须同时锁定交付物时长与排除项 20260425161344000
最短回款闭环的唯一执行链 20260425160815000
首轮小样本实验的成功与切换判据要看红线、聚集信号和下一步 20260425154734000
终点先行的检索约束 20260425154711000
ER stress UPR not yet baseline-positive as a mammalian longevity node 20260425154643000
固定范围服务的交付验收与回款闭环 20260425154639000
外联触达批次应由响应容量反推 20260425154608000
20 个潜客的周内触达应前置高温并保留响应缓冲 20260425154518000
先判红线，再看样本，最后看下一步 20260425154415000
首轮实验失败/切换触发条件要按红线、样本、下一步三层判定 20260425154400000
Heat-shock/chaperone augmentation is not yet baseline-positive as a mammalian longevity node 20260425154317000
首轮实验结果分流矩阵：先硬停，再微调，后继续，最后切换 20260425154307000
首轮实验的切换/终止阈值要与成功阈值联动 20260425154212000
Proteasome augmentation is not yet baseline-positive as a longevity node 20260425154052000
首轮小样本试验应以聚集信号与推进速度判定成败 20260425154033000
小样本首轮试验不要把单个高信号当成功 20260425153914000
不要把整类机制一刀切地排除 20260425153833000
Mitochondrial interventions have direct lifespan-endpoint positives, but evidence is heterogeneous 20260425153808000
拖款、争议、变更必须三通道分治 20260425153709000
拖款、争议与超范围变更的最小处置规则 20260425153630000
Autophagy enhancement is not yet baseline-positive like rapamycin/mTOR 20260425153601000
默认验收要有三件套 20260425153546000
里程碑应对齐成本前置点 20260425153330000
交付证据与签收动作要把默认验收变成可执行清单 20260425153228000
Rapamycin is a baseline-positive specific node; autophagy and mitochondria are not automatically longevity candidates 20260425153226000
回款先于大成本 20260425153149000
里程碑、预付款与尾款应按回款先于大成本来设计 20260425153106000
候选筛选要用“干预 + 终点”双约束检索 20260425152851000
mTOR/rapamycin 优先于泛化自噬/线粒体作为长寿候选 20260425152830000
验收默认通过，争议与变更分离 20260425152826000
验收应默认通过，争议与变更分离 20260425152806000
先算响应容量，再定批次大小 20260425152609000
批次大小应由当日响应容量反推 20260425152550000
跟进记录、状态机与复盘指标要一体化设计 20260425152400000
IL-11 的寿命/健康寿命效应大小与证据边界 20260425152333000
跟进流程要同时闭合节奏目标与判定三层 20260425152320000
跟进复盘要看高信号回复率与推进速度 20260425152236000
跟进状态机要把无回复也当作状态 20260425152215000
最小跟踪字段必须覆盖五类决策 20260425152142000
IL-11 抑制的成年哺乳动物证据层级 20260425152029000
首触达设计的三层整合：渠道、话术、判定 20260425152002000
转化条件、停止条件与切换规则的最小判定顺序 20260425151831000
跟进轮次要按唯一目标分工 20260425151755000
5次以内跟进的唯一目标分工 20260425151736000
IL-11 修复代价的直接成人哺乳动物原始反例仍未检出 20260425151726000
跟进节奏要按来源温度分层并设停止线 20260425151710000
5次以内跟进的默认节奏与停止规则 20260425151655000
首触达A/B试验的有效性判定规则 20260425151451000
Adult-mammal IL-11 inhibition: direct bone-homeostasis harm, fracture/wound evidence still missing 20260425151443000
评估首触达要看质量与推进速度 20260425151339000
首触达话术与渠道应看回复质量和推进效率 20260425151323000
记录只保留能驱动下一步决策的字段 20260425151242000
首触达后的最小跟踪字段与记录规范 20260425151220000
首触达必须先定责任人和SLA 20260425151117000
首触达响应时限、责任人与升级规则 20260425151039000
首触达回复状态的即时动作表 20260425150836000
首触达后先压缩成最小状态机 20260425150707000
IL-11 longevity benefit must be decomposed into immune/inflammatory and tissue-tradeoff components 20260425150607000
按来源分组确定首触达渠道优先级 20260425150504000
首触达只做最小可回复请求 20260425150427000
首触达话术应短、具体、低压力且只问一个问题 20260425150406000
IL-11 currently ranks above other screened immune/inflammation axes because it has direct mammalian healthspan/lifespan evidence 20260425150330000
首批潜客优先抽样暖关系和短决策链 20260425150153000
补体级联仍未跨过通用长寿瓶颈门槛 20260425150146000
前 20 个潜客应按响应概率与到对话时间优先抽样 20260425150140000
最小可交付范围要同时写清交付物与排除项 20260425145956000
最小可交付范围必须同时写清交付物与排除项 20260425145936000
IL-1/MyD88 outranks TNF/IL-6 as an upstream screening candidate, but not as a universal bottleneck 20260425145846000
首个可售 offer 应优先做成固定范围的诊断型产品化服务 20260425145704000
Complement cascade remains a high-potential but not yet universal longevity bottleneck 20260425145530000
复盘与资本再投资的一页式判定顺序 20260425145459000
复盘规则必须把“switch”做成可判定条件 20260425145419000
TNF/IL-6 are low-priority control axes in longevity bottleneck screening 20260425145355000
cGAS-STING is a strong but context-dependent aging node, not yet a universal longevity bottleneck 20260425145335000
复盘模板先验收决策能力 20260425145315000
复盘模板的最小验收标准 20260425145249000
周/月复盘共用一张骨架 20260425145229000
type I interferon / JAK-STAT is an upstream amplifier, not a default universal longevity bottleneck 20260425145224000
周/月复盘共用模板与输出格式 20260425145213000
每月复盘的固定节奏 20260425145117000
按决策半径分层设计复盘 20260425145055000
每周与每月复盘的固定节奏 20260425145036000
先保命，再加码，最后才切换 20260425144947000
复盘后的决策分流应按优先级而非平均权重处理 20260425144931000
先设安全垫，再决定加码节奏 20260425144852000
现金流与资本再投资阈值 20260425144836000
复盘只盯能驱动资本分配的少数指标 20260425144806000
赚钱复盘的核心指标面板 20260425144746000
低成本赚钱实验应优先选学习快、首单高、回款快的方案 20260425144645000
Thymic source restoration is a strong immune upstream entry, but not yet a universal longevity bottleneck 20260425144604000
资源盘点的完成判据与最小输入表 20260425144533000
资源盘点要经得起分流检验 20260425144426000
资源盘点的审稿人检验 20260425144327000
资源盘点先看是否足以驱动实验 20260425144248000
TNF/IL-6 are context-specific inflammaging mediators, not universal longevity bottlenecks 20260425144110000
先把资源约束映射成实验类型 20260425144011000
资源/约束到首轮实验的映射规则 20260425143949000
硬约束先于路径排序 20260425143901000
硬约束优先于路径偏好 20260425143838000
cGAS-STING is an upstream inflammaging amplifier, but not yet a universal longevity bottleneck 20260425143829000
先盘点资源与约束，再判断赚钱路径 20260425143735000
NLRP3 is a strong aging node but still not clearly a universal longevity bottleneck 20260425143408000
四类收入引擎的行动排序 20260425143248000
资源撮合更适合做放大器 20260425143210000
complement 轴是高潜力候选但尚非 universal longevity bottleneck 20260425143032000
交易要先覆盖波动与摩擦 20260425143008000
投资交易的可持续性取决于边际优势是否覆盖波动与摩擦 20260425142949000
创业业务的风险调整期望值通常低于现金流引擎 20260425142837000
IL-11 longevity candidacy must be split by tissue-specific tradeoffs 20260425142558000
用随机权重检验排序稳健性 20260425142558000
按风险调整期望值排赚钱路径 20260425142317000
IL-1 is a strong upstream hematopoietic inflammaging loop, but not yet a universal longevity bottleneck 20260425142133000
成年啮齿类骨折愈合中的 IL11/IL11RA1 直接因果证据仍未检出 20260425140608000
IL-11 骨折愈合原始因果证据仍未检出 20260425140012000
先找原始因果，不用综述代替证据 20260425135741000
IL-11 inhibition: bone-homeostasis evidence is direct, fracture-healing evidence remains missing 20260425135719000
Adult-mammal IL-11 inhibition has clear bone-homeostasis consequences, but direct fracture/wound inhibition evidence is sparse 20260425135357000
IL-11 still needs a pleiotropy tradeoff audit before universal-bottleneck promotion 20260425134905000
IL-6 trans-signaling is a plausible inflammaging axis, but not a universal longevity bottleneck 20260425134712000
IgG/FcRn is a strong immune age-amplifying axis, but not yet a universal longevity bottleneck 20260425134301000
IL-1 outranks TNF and type I IFN as an upstream immune loop, but still falls short of a universal longevity bottleneck 20260425133745000
IL-11 is a strong candidate but not yet a universal bottleneck 20260425133514000
反证优先于升级候选 20260425133423000
IL-17 / Th17 is a context-dependent inflammaging axis, not a universal longevity bottleneck 20260425133233000
IL-11 remains the current top immune/inflammation longevity candidate, but not yet a universal bottleneck 20260425132624000
C1q brain aging mechanism is informative but still insufficient for universal longevity candidacy 20260425131714000
寿命终点优先于通路热度 20260425131208000
IL-11 is currently the strongest immune/inflammation longevity candidate in the screening set 20260425131144000
IgG is a strong age-amplifying node, but not yet a universal longevity bottleneck 20260425130909000
Antibody / IgG accumulation is a strong age-amplifying node, but not yet a universal longevity bottleneck 20260425130643000
trained immunity is a functional defense amplifier, not a higher-priority longevity bottleneck than thymic source restoration 20260425130412000
IL-1 / emergency myelopoiesis is upstream of aged HSC bias, but not yet a universal longevity bottleneck 20260425125943000
type I interferon / JAK-STAT chronic activation is strong but context-dependent, not yet a universal bottleneck 20260425125646000
C1q is aging-associated but not yet a universal longevity bottleneck 20260425125313000
IL-11 is a strong lifespan-positive inflammaging node, but still not yet a universal immune bottleneck 20260425124410000
FGF21 不是通用免疫长寿瓶颈，而是局部胸腺 niche 入口 20260425123837000
NLRP3 is context-dependent longevity-positive, not yet a universal bottleneck 20260425051109000
单组织寿命正例不足以把炎症轴抬升为通用长寿瓶颈 20260425050954000
按因果广度而不是表面上游性来排长寿瓶颈 20260425050740000
HSC/髓系偏置尚不能压过胸腺源恢复的免疫长寿排序 20260425050727000
局部语境决定抗衰老信号是否成立 20260425050544000
Source-restoration immune entries outrank peripheral T-cell quality control when only the former has healthspan extension 20260425050521000
Mature T-cell quality control usually ranks below thymic source restoration as a longevity entry 20260425050406000
免疫长寿入口的上游性排序 20260425050322000
免疫长寿入口的上游性排序：补给源 > 外周质量控制 > 局部 niche 维护 20260425050257000
Peripheral T-cell quality control can reach frailty, but not yet lifespan 20260425050159000
Peripheral T-cell quality control can reach frailty, while thymic regeneration remains mainly immune-readout level 20260425044850000
T-cell Bcl-xL is a concrete peripheral quality-control example, but still not a validated universal longevity bottleneck 20260425044154000
Peripheral lymph node niche repair is a lower-rank immune rejuvenation entry than thymic regeneration 20260425043703000
Peripheral lymph node niche is a reversible immune entry, but not yet higher-leverage than thymic involution 20260425043458000
Thymic involution remains the strongest reversible immune entry, but not yet a validated longevity bottleneck 20260425042531000
IL-1 emergency myelopoiesis is upstream but not yet validated as a longevity bottleneck 20260425042245000
Trained immunity is not yet a validated longevity bottleneck 20260425042105000
把 inflammaging 当作回路读出而非单一瓶颈 20260425041935000
Inflammaging is better treated as a heterogeneous readout than a universal upstream bottleneck 20260425041919000
Thymus regeneration improves immune readouts but not yet lifespan evidence 20260425041751000
Immune rejuvenation can improve systemic aging phenotypes via thymic or peripheral routes 20260425041629000
Immune interventions can improve systemic aging phenotypes, but not always via thymic regeneration 20260425041447000
Immune longevity candidate ranking: thymic involution first, readouts later 20260425041227000
优先寻找“可逆上游入口”而不是只看年龄读出 20260425041000000
Thymic involution can be experimentally reversed to restore naive T-cell output in aged mice 20260425040945000
Inflammaging 更应默认按 context-dependent readout 处理 20260425040831000
Netrin-1 可作为老化骨髓 niche 的可逆上游杠杆 20260425040329000
Blood-aging candidate ranking: HSC exhaustion over CHIP; cell competition as framework 20260425040156000
先把 cell competition 当作框架，而不是立刻当作靶点 20260425040109000
老化干预要先找反馈回路 20260425035910000
Bone marrow niche rejuvenation can restore aged HSC fitness 20260425035840000
年龄相关克隆现象优先视为读出而非总瓶颈 20260425035640000
CHIP 更适合作为 HSC aging 的读出而非单一上游瓶颈 20260425035624000
CDKN2A-OSK 在皮肤创面中的 FOSL1/迁移信号更应默认按间接效应解释 20260425035159000
CDKN2A-OSK 在人皮肤创面中的直接效应与微环境间接效应应分层判断 20260425034935000
FOSL1 不能被当成 CDKN2A-OSK 直接驱动角质形成细胞迁移的证据 20260425034652000
CDKN2A-OSK 在人皮肤伤口边缘更像编辑炎症/应激状态，而非直接的增殖开关 20260425034427000
重编程效应要拆成炎症编辑与增殖激活两条轴 20260425034206000
partial reprogramming 在人皮肤修复中更像炎症/应激状态编辑，而非增殖开关 20260425034137000
partial reprogramming 在皮肤修复中更像炎症/应激状态编辑 20260425033952000
把干预时窗对齐到修复亚阶段 20260425033646000
human skin wound-healing roadmap cleanly separates acute-support and chronic-failure programs 20260425033600000
修复失败常是跨细胞协同失配 20260425033050000
chronic wound failure programs clarify which senescence states are pathological 20260425033027000
先锁定同组织的阶段-细胞分辨图谱 20260425032914000
acute skin wound healing contains distinct senescent subpopulations with opposite effects 20260425032856000
acute cutaneous wound healing is a clear example of context-dependent senescence 20260425032612000
senolysis 可能破坏有益的再生 niche 20260425032437000
partial reprogramming can improve regeneration by remodeling the extrinsic niche 20260425032418000
senolysis before OSK may remove a pro-reprogramming niche in some contexts 20260425032136000
同组织 senolytic 预处理后再做 OSK 的原始实验仍未检出 20260425031957000
senolytic 预处理与 OSK 增效证据未检出 20260425031913000
体外与器官类系统中也未检出 senolytic→OSK 原始顺序实验 20260425031655000
另一单一组织中的 senolytic→OSK 顺序组合仍未检出 20260425031525000
组合干预必须同时对齐顺序、组织和读出 20260425031020000
顺序联合比单药 head-to-head 更有信息量 20260425030652000
Topical ABT-263 treatment reduces aged skin senescence and improves subsequent wound healing 20260425030358000
皮肤 OSK vs ABT-263 参数对齐表（补全版） 20260425030211000
皮肤 OSK vs ABT-263 可执行参数表（反证版） 20260425025941000
反证式审稿的收敛规则：比较表只能在证据边界内下结论 20260425025750000
OSK vs ABT-263 的对齐/不对齐清单（皮肤场景） 20260425025615000
PMID 未命中时先回到题名和全文 20260425025540000
Topical ABT-263 in aged mouse skin: original study and readouts 20260425025523000
路由匹配的对照实验比“同剂量”更关键 20260425023023000
比较干预时必须匹配给药路线 20260425022914000
无 head-to-head 就不要做路线优劣结论 20260425022753000
OSK vs senolytic direct head-to-head remains sparse across tissues 20260425022737000
皮肤中 OSK vs ABT-263 的原始 head-to-head 未检到 20260425022429000
直接 head-to-head 证据与并列正例的判定边界 20260425022300000
同组织头对头证据比并列正例稀缺 20260425021819000
统一比较组织的选择原则：皮肤是当前可用候选面板 20260425021710000
先做同组织头对头，再谈路线优劣 20260425021502000
局部正例不能直接外推为长寿主路线代表性 20260425021446000
同一读出面板下比较长寿瓶颈候选的工程准则 20260425021221000
第二个长寿瓶颈候选的筛选原则 20260425021002000
重编程验证必须用三重读出 20260425020521000
单一衰老读出不够，必须配安全反证读出 20260425015958000
先锁定单一可验证瓶颈，再谈宏大目标 20260425015659000
先随机后穷举地压力测试判别协议 20260425015357000
三态判别协议的压力测试结果 20260425015339000
旧前缀+下一前缀联合生存刻画中性（small-case） 20260425014335000
旧前缀可替代嵌入不足以判别中性 20260425013950000
候选影响区不等于敏感区 20260425012422000
锚点固定后先局部化，再做单侧交换 20260425011113000
左侧一次相邻交换的切分判定复用模板 20260425005439000
先找切分点，再谈独立性 20260425002443000
局部化不等于可分解 20260424235814000
Iter-4360dd15-0184-failure-left-split-overclaim 20260424235623000
锚定后交换只改一侧 20260424235519000
用 fresh marker 规范化 witness 20260424235110000
先定可判定窗口再谈可达性 20260424234936000
先把可判定条件和边界对齐 20260424234753000
先定语义再证充要 20260424234637000
先锁定支持集，再谈窗口可达性 20260424234434000
先做可达性穷举对账 20260424234150000
先证支持集，再谈窗口判定 20260424234005000
先证窗口等价，再谈判定完备性 20260424233824000
窗口语义要显式区分 20260424233737000
窗口语义要区分锚点距离与支持集包含 20260424233537000
先锁定操作索引再谈窗口 20260424233448000
交换必须受窗口约束 20260424233043000
纯插入与轻微重排要分开判 20260424232542000
先保骨架再谈编辑量 20260424232439000
先判定纠错类型再谈规则 20260424232347000
Iter-4360dd15-0164-lesson-v2-rule-false-positives 20260424232336000
二层规则先看强度再看重叠 20260424232219000
重写判定要分层 20260424231903000
重写判定要用二层信号 20260424231655000
重写判定必须分层 20260424231556000
先定位实体，再抽规则 20260424231410000
Iter-4360dd15-0159-lesson-rewrite-threshold-failure 20260424231322000
先验证模式，再封装判定 20260424231142000
先验证事实，再封装工具 20260424231044000
先拿到原始 XML 再做差分 20260424230933000
先取 XML 再谈差分 20260424230817000
先拿原文再做逐词 diff 20260424230659000
摘要不等于证据 20260424230541000
先拿到原文再判定改动 20260424230336000
先拿正文再判类型 20260424230205000
先抓正文再判定改动类型 20260424225933000
Iter-4360dd15-0149-lesson-pmc4083033-not-insertion 20260424225824000
先锁定可验证候选，再做差分 20260424225616000
先锁定勘误页，再做最小差分 20260424225328000
先找确定样本再测分类器 20260424224859000
先做原句对齐再判定修正类型 20260424224634000
先锁定原句再做最小差分 20260424224537000
先固定原文再做差分 20260424224442000
先锁权威源再逐字对照 20260424223604000
先锁定权威源再逐字抽句 20260424222955000
先锁定权威源再逐字对照 20260424222844000
先抓原句边界再谈纠错范围 20260424222053000
先抓纠错点再补全文本 20260424221855000
先锁定原句，再做逐字差分 20260424221748000
先取原句，再判修订边界 20260424221550000
先拿原句再判修订边界 20260424221302000
先取原句再认修订 20260424221033000
BOLERO-2 erratum exact abstract sentence and boundary of correction 20260424221021000
先锁定原文再做修订对照 20260424220853000
先判更正范围再定引用文本 20260424220712000
更正只改被点名的那一句 20260424220536000
先拿到更正全文再判断影响 20260424220425000
先锁书目信息，再查更正 20260424220217000
先锁主文再抽全字段 20260424220056000
找到主文后一次性抽全字段 20260424215932000
先锁定可读全文，再提取数值 20260424215754000
先抓可访问摘要再补全文闭环 20260424215304000
先用可访问二源闭环再碰受限原文 20260424215113000
题录线索也要做双源闭环 20260424214532000
先判定证据层级，再找精确 PMID 20260424214413000
亚组论文不能当原始试验 20260424214200000
别在记忆里绕圈，直接打原始证据 20260424214025000
先判研究类型，再谈是否相关 20260424213803000
优先走可解析的证据链 20260424213650000
先验明响应格式再写解析器 20260424213353000
先拿全量参考文献再分类 20260424212846000
先拿参考文献清单再做分类 20260424212541000
先抓参考文献，再做证据分层 20260424212326000
Iter-4360dd15-0099-critic-mechanistic-vs-aging-trial 20260424212312000
别从落地页取证 20260424212107000
用可访问的镜像入口绕开 403 20260424211826000
先拿到可访问的全文入口 20260424211456000
先从综述参考链落到单个原始 PMID 20260424211152000
先从参考链锁定单个原始试验 20260424210848000
先从综述引用链提取唯一原始 PMID 20260424210659000
先锁定唯一 PMID 再判终点 20260424210407000
先锁定唯一引用再看终点 20260424210210000
先从综述引用链落到原始试验 20260424210023000
先从综述引用链下钻到原始试验 20260424205810000
PMID 31761958 is skin-senescence endpoint, not longevity evidence 20260424205759000
先锁定原始试验再谈终点 20260424205610000
综述只负责引路 20260424205350000
先证终点，再谈长寿 20260424204906000
线索要尽快落到单个 PMID 20260424204736000
综述只能当入口，不能当证据 20260424204557000
先锁原始试验再谈长寿终点 20260424203542000
PEARL trial is not direct longevity evidence 20260424203525000
先用终点语境判定是否真是长寿证据 20260424203253000
先最小化候选，再三字段补全 20260424203148000
先用注册库补关键字段 20260424203004000
先抽字段再分类 20260424202727000
先做摘要级字段抽取 20260424202425000
先确认研究类型再谈终点 20260424202124000
先抽字段再定规则 20260424201859000
先表格化再抽象规则 20260424201631000
先用现成抓取验证，再封装工具 20260424201421000
优先用可访问全文源替代出版社页 20260424201155000
绕开 PubMed HTML，优先用 PMC/Europe PMC 抽摘要 20260424200940000
先用题名证据锁定研究性质 20260424200615000
先确保可输出再做细化 20260424200233000
先拿原始候选表再分类 20260424200123000
边界术语先做例外判定 20260424195952000
先分类型再写表 20260424195838000
先定审计表再搜证 20260424195730000
先定原始性再逐条填表 20260424195602000
先判原始性再补 PMID 20260424195432000
Longitudinal analysis is not an automatic exclusion in trial titles 20260424195414000
先用题名锁定原始性 20260424195244000
先定研究原始性再补 PMID 20260424194952000
先排除二次分析再做原始性归类 20260424194723000
先判定研究原始性再做保留表 20260424194535000
先把候选映射到具体研究类型 20260424194357000
先排除非原始研究 20260424194237000
Iter-4360dd15-0049-crossref-candidate-overgeneration-critique 20260424194224000
先拿可读副本再抽取 20260424194120000
先确认可读副本再抽取 20260424193842000
先拿到可读副本再做抽取 20260424193558000
PMID 38310895 access topology verified 20260424193545000
先绕过访问阻断再做抽取 20260424193149000
先拿到表，再做映射 20260424192851000
候选召回不能替代纳入判定 20260424192629000
候选召回不能代替纳入判定 20260424192437000
先用可机读引用链缩窄候选，再补全文表 20260424191824000
Critique of Crossref trialish filtering for PMID 38310895 reference chain 20260424191745000
先拿引用数组再做精筛 20260424191630000
先用 Crossref 参考数组替代网页正文 20260424191122000
先找机器可读源 20260424190238000
先拿到可解析的参考文献源 20260424190005000
先沿参考文献链找可抓取入口 20260424185749000
先用可访问元数据逼近全文表格 20260424185413000
先锁具体入口，再谈全文 20260424185116000
先用题名+DOI锁入口 20260424184550000
先找可读镜像再抽表 20260424184205000
先锁定可读全文入口 20260424183902000
先绕开 403 再抽表 20260424183147000
先拿可访问全文入口 20260424182646000
先找可抓取全文入口 20260424182104000
先用可访问全文定位表格再抽取 20260424181640000
先用可访问入口锁定表格 20260424181027000
先找可访问的二级证据入口 20260424180750000
先拿到可访问全文表格 20260424180508000
先锁定可访问综述表格再补全缺口 20260424180159000
先用可访问综述骨架锁定缺口 20260424175843000
2024 rapamycin systematic review: abstract-level constraints and what remains missing 20260424175824000
先用可访问综述骨架再补缺项 20260424175603000
先锁定综述正文里的表，再追原始试验 20260424175236000
先用综述表做骨架再回填原始试验 20260424174951000
先把异质证据拆成逐试验清单 20260424174404000
先用可访问二级来源拼表 20260424174242000
Iter-4360dd15-0004-critique-rapamycin-evidence 20260424174226000
先用摘要锁定，再补全文校验 20260424173853000
先抓人类试验证据表 20260424173558000
先选一条机制再读人类证据 20260424173249000
先压缩成最小状态机再验证 20260424171825000
先压缩再证伪 20260424171717000
先统计差值，再找模式 20260424171345000
先枚举再抽象 20260424171249000
先用枚举锁定可证规律 20260424171201000