{"id":80181,"topic":"ai","source":"Sohu","title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","url_hash":"c35c2f5c4518bcb32a2144dc2aaad833f82a18bc","author":"","summary":"<a href=\"https://news.google.com/rss/articles/CBMiiAFBVV95cUxQX0pMX00xa0J1QUZQcUZGZ1RGaUkzdk1qZ3JPdnBWY0lObnlfeTktaEw0WjJRVVJnZTN4YU1ZU2xyTU1qV2VROWZ3MTlDNDk2b1JVQ3NMUkVzbzIxaUNmQUhQbTU0UWk4MFlCNDI4RFpEYjRGbno5bzRVRGx6ZlV5aTBtVGtkNnQx?oc=5\" target=\"_blank\">奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native</a>&nbsp;&nbsp;<font color=\"#6f6f6f\">Sohu</font>","content":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI Native和应用自进化等多角度，系统化推进研发改革。技术升级是为了让现有业务获得更好的效果，更高的效率；AI Native是探索新技术带来新的增长点；应用自进化则是智能体时代的新范式，让模型能力能够在上线后具有自主学习能力，越用越好，自适应地解决实际问题。\n在技术升级方面，杨叶辉以奇富科技在视觉场景的模型改造为例。过去，奇富科技在视觉场景中分散部署了70多个模型，覆盖OCR、分类、检测和识别等任务。模型能力高度重合，却需要分别维护、独立部署，运维成本高，GPU资源利用率低。传统判别式模型还存在推理和泛化天花板：面对银行流水、合同等复杂材料能力不足；每增加一种证件，往往都要重新采集数据、训练模型。","image_url":"//q4.itc.cn/q_70/images03/20260911/7d71fd0f9b5447378c85baf5e62fbfd0.jpeg","lang":"zh","published_at":"2026-09-11T10:25:45+00:00","fetched_at":"2026-09-11T11:15:08+00:00","status":"read","starred":0,"extract_state":"ok","summary_auto":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","cluster_id":null,"extract_retries":0,"extract_error":null,"contract_version":"news_item.v1","format_contract_version":"news_item_formats.v1","dedup_url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 482 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":482,"summary_length":180,"usable_text_length":482,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":482,"summary_length":180}},"news_item":{"id":80181,"canonical_url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","source_url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","source_name":"Sohu","author":null,"published_at":"2026-09-11T10:25:45+00:00","locale":"zh","topic":"ai","tags":[],"rss_summary":"<a href=\"https://news.google.com/rss/articles/CBMiiAFBVV95cUxQX0pMX00xa0J1QUZQcUZGZ1RGaUkzdk1qZ3JPdnBWY0lObnlfeTktaEw0WjJRVVJnZTN4YU1ZU2xyTU1qV2VROWZ3MTlDNDk2b1JVQ3NMUkVzbzIxaUNmQUhQbTU0UWk4MFlCNDI4RFpEYjRGbno5bzRVRGx6ZlV5aTBtVGtkNnQx?oc=5\" target=\"_blank\">奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native</a>&nbsp;&nbsp;<font color=\"#6f6f6f\">Sohu</font>","full_text":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI Native和应用自进化等多角度，系统化推进研发改革。技术升级是为了让现有业务获得更好的效果，更高的效率；AI Native是探索新技术带来新的增长点；应用自进化则是智能体时代的新范式，让模型能力能够在上线后具有自主学习能力，越用越好，自适应地解决实际问题。\n在技术升级方面，杨叶辉以奇富科技在视觉场景的模型改造为例。过去，奇富科技在视觉场景中分散部署了70多个模型，覆盖OCR、分类、检测和识别等任务。模型能力高度重合，却需要分别维护、独立部署，运维成本高，GPU资源利用率低。传统判别式模型还存在推理和泛化天花板：面对银行流水、合同等复杂材料能力不足；每增加一种证件，往往都要重新采集数据、训练模型。","excerpt":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 482 characters.","diagnostics_url":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 482 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":482,"summary_length":180,"usable_text_length":482,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":482,"summary_length":180}}},"display_formats":["compact","card","full","digest_section","json"]},"daily_stack_record":{"title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","summary":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","source":"Sohu","date":"2026-09-11T10:25:45+00:00","content":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI Native和应用自进化等多角度，系统化推进研发改革。技术升级是为了让现有业务获得更好的效果，更高的效率；AI Native是探索新技术带来新的增长点；应用自进化则是智能体时代的新范式，让模型能力能够在上线后具有自主学习能力，越用越好，自适应地解决实际问题。\n在技术升级方面，杨叶辉以奇富科技在视觉场景的模型改造为例。过去，奇富科技在视觉场景中分散部署了70多个模型，覆盖OCR、分类、检测和识别等任务。模型能力高度重合，却需要分别维护、独立部署，运维成本高，GPU资源利用率低。传统判别式模型还存在推理和泛化天花板：面对银行流水、合同等复杂材料能力不足；每增加一种证件，往往都要重新采集数据、训练模型。","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 482 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 482 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":482,"summary_length":180,"usable_text_length":482,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":482,"summary_length":180}},"tags":[]},"fallback_formats":["markdown","json","html"],"actions":{"read":"/item/80181","export_markdown":"/api/items/80181/export?format=markdown","export_json":"/api/items/80181/export?format=json","diagnose":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334"},"formats":{"full":{"id":80181,"title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","source":"Sohu","author":null,"published_at":"2026-09-11T10:25:45+00:00","locale":"zh","topic":"ai","tags":[],"excerpt":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","full_text":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI Native和应用自进化等多角度，系统化推进研发改革。技术升级是为了让现有业务获得更好的效果，更高的效率；AI Native是探索新技术带来新的增长点；应用自进化则是智能体时代的新范式，让模型能力能够在上线后具有自主学习能力，越用越好，自适应地解决实际问题。\n在技术升级方面，杨叶辉以奇富科技在视觉场景的模型改造为例。过去，奇富科技在视觉场景中分散部署了70多个模型，覆盖OCR、分类、检测和识别等任务。模型能力高度重合，却需要分别维护、独立部署，运维成本高，GPU资源利用率低。传统判别式模型还存在推理和泛化天花板：面对银行流水、合同等复杂材料能力不足；每增加一种证件，往往都要重新采集数据、训练模型。","reading_time_min":1,"extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 482 characters.","diagnostics_url":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 482 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":482,"summary_length":180,"usable_text_length":482,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":482,"summary_length":180}}},"quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 482 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":482,"summary_length":180,"usable_text_length":482,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":482,"summary_length":180}},"actions":{"read":"/item/80181","export_markdown":"/api/items/80181/export?format=markdown","export_json":"/api/items/80181/export?format=json","diagnose":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334"}},"digest":{"id":80181,"title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","source":"Sohu","topic":"ai","published_at":"2026-09-11T10:25:45+00:00","excerpt":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native 9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。 在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","quality_bucket":"high","quality_reason":"High confidence: full text extraction produced 482 characters.","reading_time_min":1,"cluster_id":null},"card":{"display_title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","subtitle":"Sohu · 2026-09-11","summary":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native 9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。 在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","badges":["quality:high"],"links":{"read":"/item/80181","original":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","diagnose":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334"},"quality_warning":null},"export":{"title":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native - Sohu","url":"https://m.sohu.com/a/1074883460_211762?scm=10001.325_13-325_13.0.0-0-0-0-0.5_1334","summary":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI…","source":"Sohu","date":"2026-09-11T10:25:45+00:00","content":"奇富科技杨叶辉：以大模型升级金融AI底座，加速迈向AI Native\n9月11日消息，在近日的2026年服贸会专题活动“中国AIGC创新应用交流会”上，奇富科技多模态负责人杨叶辉博士分享公司以统一多模态大模型底座推动金融业务升级、加速AI Native转型的实践。\n在奇富科技看来，金融AI的关键突破，不是增加多少大模型嵌入的工作流，而是要按照技术升级、AI Native和应用自进化等多角度，系统化推进研发改革。技术升级是为了让现有业务获得更好的效果，更高的效率；AI Native是探索新技术带来新的增长点；应用自进化则是智能体时代的新范式，让模型能力能够在上线后具有自主学习能力，越用越好，自适应地解决实际问题。\n在技术升级方面，杨叶辉以奇富科技在视觉场景的模型改造为例。过去，奇富科技在视觉场景中分散部署了70多个模型，覆盖OCR、分类、检测和识别等任务。模型能力高度重合，却需要分别维护、独立部署，运维成本高，GPU资源利用率低。传统判别式模型还存在推理和泛化天花板：面对银行流水、合同等复杂材料能力不足；每增加一种证件，往往都要重新采集数据、训练模型。","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//m.sohu.com/a/1074883460_211762%3Fscm%3D10001.325_13-325_13.0.0-0-0-0-0.5_1334","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 482 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 482 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":482,"summary_length":180,"usable_text_length":482,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":482,"summary_length":180}},"tags":[],"format_contract_version":"news_item_formats.v1"}}}