{"id":1191813,"date":"2026-10-28T20:03:48","date_gmt":"2026-10-29T03:03:48","guid":{"rendered":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/?post_type=msr-job-opportunity&p=1191813"},"modified":"2026-10-28T20:17:29","modified_gmt":"2026-10-29T03:17:29","slug":"research-intern-for-machine-learning-group-discriminative-foundation-models-and-efficient-decision-making","status":"publish","type":"msr-job-opportunity","link":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/opportunity\/research-intern-for-machine-learning-group-discriminative-foundation-models-and-efficient-decision-making\/","title":{"rendered":"Research Intern for Machine Learning Group \u2013 Discriminative Foundation Models and Efficient Decision-Making\u00a0"},"content":{"rendered":"\n\n\n

Position Title:<\/strong> Research Intern for Machine Learning Group \u2013 Discriminative Foundation Models and Efficient Decision-Making <\/p>\n\n\n\n

Job Type:<\/strong> Full-time Intern 
Number of Openings:<\/strong> 2-4 
Location:<\/strong> Beijing <\/p>\n\n\n\n

Group Introduction:<\/strong> <\/p>\n\n\n\n

We study intelligent agents that can perceive, reason, plan, and act in complex interactive environments. Our research brings together decision models, large language models, multimodal models, and reinforcement learning, enabling agents to learn continuously from interaction and improve their decision-making capabilities. Over the past few years, we have published numerous papers in leading international conferences and journals. We also have extensive expertise in reinforcement learning: our Mahjong AI system, Suphx, became the first AI system to achieve the rank of ten-dan on the internationally renowned competitive Mahjong platform Tenhou, surpassing the average performance of top human players in the platform\u2019s public rooms. <\/p>\n\n\n\n

This position focuses on Decision Foundation Models for discriminative tasks\u2014models that turn natural-language task descriptions and input context into fast, structured decisions that software can use directly.<\/strong> Rather than generating free-form text, these models select among specified options, make binary judgments, or assign scores, with an emphasis on well-calibrated probabilities and meaningful uncertainty estimates. We explore large-scale multitask learning, reinforcement learning for calibrated decisions, data synthesis and selection, lightweight architectures, and efficient inference. Our goal is to build reusable decision-making capabilities that generalize across tasks while balancing accuracy, reliability, latency, and cost for agents and automated workflows. <\/p>\n\n\n\n

Job Responsibilities:<\/strong> <\/p>\n\n\n\n

Contribute to the research and development of\u202fDecision Foundation Models<\/strong>, building on the team\u2019s existing training framework. Investigate data, learning objectives, model architectures, and inference methods to improve performance across diverse discriminative tasks. Specific responsibilities include: <\/p>\n\n\n\n

1. Training and Post-Training Methods<\/strong>:\u202fInvestigate multitask learning and reinforcement learning objectives for accurate, well-calibrated decisions, including research directions motivated by Reinforcement Learning for Calibrated Decisions (RLCD). Study how to jointly improve predictive accuracy, probability calibration, and uncertainty estimation. <\/p>\n\n\n\n

2. Data Development and Optimization<\/strong>:\u202fCollect, curate, and synthesize training data for classification, option selection, binary judgment, and rubric-based scoring. Explore data quality assessment, selection, and task-mixture strategies to improve data efficiency and generalization to new task descriptions and label sets. <\/p>\n\n\n\n

3. Lightweight Architectures and Efficient Inference:<\/strong>\u202fExplore architectures and prediction mechanisms tailored to structured decision outputs, alongside knowledge distillation and inference optimization. Reduce latency, GPU memory usage, and serving costs while preserving predictive quality. <\/p>\n\n\n\n

4. Evaluation and Research Iteration<\/strong>:\u202fDevelop reproducible evaluations across domains, unseen tasks, and distribution shifts. Assess accuracy, calibration, robustness, and inference efficiency, including how uncertainty estimates support abstention or escalation to human review. Identify research questions through systematic experiments and failure analysis. <\/p>\n\n\n\n

Qualifications:<\/strong> <\/p>\n\n\n\n

    \n
  1. PhD in Computer Science, Software Engineering, Electrical Engineering, or other related fields. Undergraduate and Master\u2019s students aspiring to scientific research are also welcome to apply.\u00a0<\/li>\n<\/ol>\n\n\n\n
      \n
    1. Strong Python programming skills and proficiency in using AI coding tools to accelerate research and development, with the ability to understand, validate, and improve generated code. Good communication and teamwork skills.\u00a0<\/li>\n<\/ol>\n\n\n\n
        \n
      1. A solid foundation in machine learning and deep learning, familiarity with frameworks such as PyTorch, and the ability to read current research papers and conduct experiments. A strong interest in discriminative foundation models, structured prediction, calibrated decision-making, and efficient inference.\u00a0<\/li>\n<\/ol>\n\n\n\n
          \n
        1. Bonus points: Familiarity or hands-on experience with models designed for structured decision-making, such as Jev; research experience in classification and scoring models, reinforcement learning, probability calibration and uncertainty estimation, synthetic data generation and data selection, model distillation, or lightweight architectures; achievements in mathematics, physics, or informatics competitions; or publications in top international conferences and journals..\u00a0<\/li>\n<\/ol>\n\n\n\n

          Internship Duration Requirements:<\/strong> 
          Must obtain permission from your academic advisor and commit to at least three months of internship. <\/p>\n\n\n\n

          Please be sure to download and complete the application form (Application form link: https:\/\/aka.ms\/InternApplication (opens in new tab)<\/span><\/a> ) and send it along with a complete English and Chinese resume (in PDF\/Word format) to: MSRAih@microsoft.com<\/a> & lizo@microsoft.com<\/a>. Please include ” Research Intern for Machine Learning Group \u2013 Discriminative Foundation Models and Efficient Decision-Making ” in the email subject line. 
          To learn more about the \u201cStars of Tomorrow Internship Program,\u201d please visit:           https:\/\/www.msra.cn\/zh-cn\/connections\/academic-programs\/tomorrowstars (opens in new tab)<\/span><\/a> <\/p>\n\n\n\n

          <\/p>\n\n\n\n

          <\/p>\n\n\n\n

          \u5c97\u4f4d\u540d\u79f0\uff1a<\/strong>\u673a\u5668\u5b66\u4e60\u7ec4\u7814\u7a76\u5b9e\u4e60\u751f \u2013 \u5224\u522b\u5f0f\u57fa\u7840\u6a21\u578b\u4e0e\u9ad8\u6548\u51b3\u7b56\u65b9\u5411 <\/p>\n\n\n\n

          \u5de5\u4f5c\u6027\u8d28\uff1a<\/strong>\u5168\u804c\u5b9e\u4e60\u751f <\/p>\n\n\n\n

          \u62db\u8058\u4eba\u6570\uff1a<\/strong>2-4\u4eba <\/p>\n\n\n\n

          \u5de5\u4f5c\u5730\u70b9\uff1a<\/strong>\u5317\u4eac <\/p>\n\n\n\n

          \u7ec4\u522b\u7b80\u4ecb\uff1a<\/strong> <\/p>\n\n\n\n

          \u6211\u4eec\u81f4\u529b\u4e8e\u7814\u7a76\u80fd\u591f\u5728\u590d\u6742\u4ea4\u4e92\u73af\u5883\u4e2d\u611f\u77e5\u3001\u63a8\u7406\u3001\u89c4\u5212\u548c\u884c\u52a8\u7684\u667a\u80fd\u4f53\uff0c\u63a2\u7d22\u51b3\u7b56\u6a21\u578b\u4e0e\u5927\u8bed\u8a00\u6a21\u578b\u3001\u591a\u6a21\u6001\u6a21\u578b\u3001\u5f3a\u5316\u5b66\u4e60\u7684\u878d\u5408\uff0c\u4f7f\u667a\u80fd\u4f53\u80fd\u591f\u901a\u8fc7\u73af\u5883\u4ea4\u4e92\u6301\u7eed\u5b66\u4e60\u5e76\u63d0\u5347\u51b3\u7b56\u80fd\u529b\u3002\u8fc7\u53bb\u51e0\u5e74\uff0c\u6211\u4eec\u5728\u56fd\u9645\u9876\u7ea7\u4f1a\u8bae\u548c\u671f\u520a\u4e0a\u53d1\u8868\u4e86\u5927\u91cf\u8bba\u6587\u3002\u6211\u4eec\u5728\u5f3a\u5316\u5b66\u4e60\u9886\u57df\u4e5f\u6709\u6df1\u539a\u79ef\u7d2f\uff1a\u6211\u4eec\u7684\u9ebb\u5c06 AI \u7cfb\u7edf Suphx \u6210\u4e3a\u9996\u4e2a\u5728\u56fd\u9645\u77e5\u540d\u4e13\u4e1a\u9ebb\u5c06\u5e73\u53f0\u201c\u5929\u51e4\u201d\u4e0a\u8fbe\u5230\u5341\u6bb5\u7684 AI \u7cfb\u7edf\uff0c\u5728\u8be5\u5e73\u53f0\u516c\u5f00\u623f\u95f4\u4e2d\u8d85\u8d8a\u4e86\u9876\u5c16\u4eba\u7c7b\u73a9\u5bb6\u7684\u5e73\u5747\u6c34\u5e73\u3002 <\/p>\n\n\n\n

          \u672c\u5c97\u4f4d\u805a\u7126\u4e8e\u9762\u5411\u5224\u522b\u5f0f\u4efb\u52a1\u7684\u51b3\u7b56\u57fa\u7840\u6a21\u578b\uff08Decision Foundation Models\uff09<\/strong>\uff0c\u4f7f\u6a21\u578b\u80fd\u591f\u6839\u636e\u4efb\u52a1\u63cf\u8ff0\u76f4\u63a5\u8f93\u51fa\u5206\u7c7b\u3001\u5224\u65ad\u3001\u9009\u62e9\u6216\u8bc4\u5206\u7ed3\u679c\uff0c\u5e76\u63d0\u4f9b\u53ef\u9760\u7684\u6982\u7387\u4e0e\u7f6e\u4fe1\u5ea6\u3002\u6211\u4eec\u63a2\u7d22\u5927\u89c4\u6a21\u591a\u4efb\u52a1\u8bad\u7ec3\u3001\u9762\u5411\u6821\u51c6\u51b3\u7b56\u7684\u5f3a\u5316\u5b66\u4e60\u3001\u6570\u636e\u5408\u6210\u4e0e\u7b5b\u9009\uff0c\u4ee5\u53ca\u8f7b\u91cf\u5316\u67b6\u6784\u4e0e\u9ad8\u6548\u63a8\u7406\uff0c\u81f4\u529b\u4e8e\u5b9e\u73b0\u8de8\u4efb\u52a1\u6cdb\u5316\u3001\u4f4e\u5ef6\u8fdf\u3001\u4f4e\u6210\u672c\u7684\u53ef\u9760\u51b3\u7b56\uff0c\u4e3a\u667a\u80fd\u4f53\u4e0e\u81ea\u52a8\u5316\u7cfb\u7edf\u63d0\u4f9b\u53ef\u590d\u7528\u7684\u57fa\u7840\u51b3\u7b56\u80fd\u529b\u3002 <\/p>\n\n\n\n

          \u5de5\u4f5c\u804c\u8d23\uff1a<\/strong> <\/p>\n\n\n\n

          \u53c2\u4e0e\u51b3\u7b56\u57fa\u7840\u6a21\u578b<\/strong>\u7684\u7814\u53d1\uff0c\u5728\u56e2\u961f\u5df2\u6709\u8bad\u7ec3\u6846\u67b6\u4e0a\u6301\u7eed\u8fed\u4ee3\uff0c\u56f4\u7ed5\u6570\u636e\u3001\u8bad\u7ec3\u65b9\u6cd5\u3001\u6a21\u578b\u7ed3\u6784\u4e0e\u63a8\u7406\u6548\u7387\u5f00\u5c55\u7814\u7a76\uff0c\u63d0\u5347\u6a21\u578b\u5728\u5e7f\u6cdb\u5224\u522b\u5f0f\u4efb\u52a1\u4e0a\u7684\u6cdb\u5316\u80fd\u529b\u3001\u9884\u6d4b\u53ef\u9760\u6027\u4e0e\u8ba1\u7b97\u6548\u7387\u3002\u5177\u4f53\u5305\u62ec\uff1a <\/p>\n\n\n\n

            \n
          1. \u8bad\u7ec3\u4e0e\u540e\u8bad\u7ec3\u7b97\u6cd5<\/strong>\uff1a\u63a2\u7d22\u591a\u4efb\u52a1\u5b66\u4e60\uff0c\u4ee5\u53ca\u9762\u5411\u6821\u51c6\u51b3\u7b56\u7684\u5f3a\u5316\u5b66\u4e60\u65b9\u6cd5\uff08\u5982 RLCD \u76f8\u5173\u65b9\u5411\uff09\uff0c\u8054\u5408\u63d0\u5347\u51b3\u7b56\u51c6\u786e\u7387\u3001\u6982\u7387\u6821\u51c6\u4e0e\u4e0d\u786e\u5b9a\u6027\u4f30\u8ba1\u80fd\u529b\u3002 <\/li>\n<\/ol>\n\n\n\n
              \n
            1. \u6570\u636e\u6784\u5efa\u4e0e\u4f18\u5316<\/strong>\uff1a\u6536\u96c6\u3001\u6e05\u6d17\u5e76\u5408\u6210\u8986\u76d6\u5206\u7c7b\u3001\u5224\u65ad\u3001\u9009\u62e9\u548c\u8bc4\u5206\u7b49\u4efb\u52a1\u7684\u8bad\u7ec3\u6570\u636e\uff0c\u63a2\u7d22\u6570\u636e\u7b5b\u9009\u3001\u8d28\u91cf\u8bc4\u4f30\u4e0e\u4efb\u52a1\u914d\u6bd4\u7b56\u7565\uff0c\u63d0\u5347\u6570\u636e\u6548\u7387\u4e0e\u4efb\u52a1\u8986\u76d6\u5ea6\u3002 <\/li>\n<\/ol>\n\n\n\n
                \n
              1. \u8f7b\u91cf\u5316\u4e0e\u9ad8\u6548\u63a8\u7406<\/strong>\uff1a\u7814\u7a76\u9002\u5408\u5224\u522b\u5f0f\u4efb\u52a1\u7684\u6a21\u578b\u7ed3\u6784\u3001\u77e5\u8bc6\u84b8\u998f\u4e0e\u63a8\u7406\u4f18\u5316\u65b9\u6cd5\uff0c\u964d\u4f4e\u5ef6\u8fdf\u3001\u663e\u5b58\u5360\u7528\u548c\u90e8\u7f72\u6210\u672c\uff0c\u4f18\u5316\u51c6\u786e\u7387\u4e0e\u6548\u7387\u4e4b\u95f4\u7684\u6743\u8861\u3002 <\/li>\n<\/ol>\n\n\n\n
                  \n
                1. \u8bc4\u6d4b\u4e0e\u7814\u7a76\u8fed\u4ee3<\/strong>\uff1a\u5efa\u7acb\u8986\u76d6\u8de8\u9886\u57df\u3001\u672a\u89c1\u4efb\u52a1\u53ca\u5206\u5e03\u53d8\u5316\u7684\u8bc4\u6d4b\u4f53\u7cfb\uff0c\u4ece\u9884\u6d4b\u8d28\u91cf\u3001\u6821\u51c6\u7a0b\u5ea6\u548c\u63a8\u7406\u6210\u672c\u7b49\u7ef4\u5ea6\u5206\u6790\u6a21\u578b\u8868\u73b0\uff0c\u63d0\u70bc\u7814\u7a76\u95ee\u9898\u5e76\u5f00\u5c55\u53ef\u590d\u73b0\u7684\u5b9e\u9a8c\u9a8c\u8bc1\u3002 <\/li>\n<\/ol>\n\n\n\n

                  \u4efb\u804c\u8981\u6c42\uff1a<\/strong> <\/p>\n\n\n\n

                  1. \u8ba1\u7b97\u673a\u79d1\u5b66\u3001\u8f6f\u4ef6\u5de5\u7a0b\u3001\u7535\u5b50\u5de5\u7a0b\u6216\u5176\u5b83\u76f8\u5173\u4e13\u4e1a\u535a\u58eb\uff0c\u6709\u5fd7\u4e8e\u79d1\u7814\u7684\u672c\u79d1\u751f\u6216\u7855\u58eb\u4e5f\u6b22\u8fce\u7533\u8bf7\uff1b <\/p>\n\n\n\n

                  2. \u5177\u5907\u624e\u5b9e\u7684 Python \u7f16\u7a0b\u80fd\u529b\uff0c\u80fd\u591f\u719f\u7ec3\u4f7f\u7528 AI coding \u5de5\u5177\u9ad8\u6548\u5f00\u5c55\u7814\u53d1\uff0c\u5e76\u5bf9\u751f\u6210\u4ee3\u7801\u8fdb\u884c\u7406\u89e3\u3001\u9a8c\u8bc1\u4e0e\u6539\u8fdb\uff1b\u5177\u5907\u826f\u597d\u7684\u6c9f\u901a\u80fd\u529b\u548c\u56e2\u961f\u534f\u4f5c\u7cbe\u795e\u3002 <\/p>\n\n\n\n

                  3. \u5177\u5907\u673a\u5668\u5b66\u4e60\u4e0e\u6df1\u5ea6\u5b66\u4e60\u57fa\u7840\uff0c\u719f\u6089 PyTorch \u7b49\u6846\u67b6\uff0c\u80fd\u591f\u9605\u8bfb\u76f8\u5173\u9886\u57df\u7684\u524d\u6cbf\u8bba\u6587\u5e76\u5f00\u5c55\u5b9e\u9a8c\uff1b\u5bf9\u5224\u522b\u5f0f\u57fa\u7840\u6a21\u578b\u3001\u7ed3\u6784\u5316\u51b3\u7b56\u3001\u6982\u7387\u6821\u51c6\u4e0e\u9ad8\u6548\u63a8\u7406\u6709\u6d53\u539a\u5174\u8da3\u3002 <\/p>\n\n\n\n

                  4. **\u52a0\u5206\u9879\uff1a**\u4e86\u89e3\u6216\u4f7f\u7528\u8fc7 Jev \u7b49\u9762\u5411\u7ed3\u6784\u5316\u51b3\u7b56\u7684\u6a21\u578b\uff1b\u5177\u6709\u5206\u7c7b\u4e0e\u8bc4\u5206\u6a21\u578b\u3001\u5f3a\u5316\u5b66\u4e60\u3001\u6982\u7387\u6821\u51c6\u4e0e\u4e0d\u786e\u5b9a\u6027\u4f30\u8ba1\u3001\u5408\u6210\u6570\u636e\u4e0e\u6570\u636e\u7b5b\u9009\u3001\u6a21\u578b\u84b8\u998f\u6216\u8f7b\u91cf\u5316\u67b6\u6784\u7b49\u76f8\u5173\u7814\u7a76\u7ecf\u9a8c\uff1b\u6709\u6570\u5b66\u3001\u7269\u7406\u6216\u4fe1\u606f\u5b66\u7ade\u8d5b\u6210\u7ee9\uff0c\u6216\u5728\u56fd\u9645\u9876\u7ea7\u4f1a\u8bae\u548c\u671f\u520a\u4e0a\u53d1\u8868\u8fc7\u8bba\u6587\u3002 <\/p>\n\n\n\n

                  \u5de5\u4f5c\u65f6\u95f4\u8981\u6c42<\/strong>\uff1a<\/p>\n\n\n\n

                  \u80fd\u83b7\u5f97\u5bfc\u5e08\u8bb8\u53ef\u5e76\u4fdd\u8bc1\u81f3\u5c11\u4e09\u4e2a\u6708\u7684\u5b9e\u4e60\u3002 <\/p>\n\n\n\n

                  \u8bf7\u52a1\u5fc5\u4e0b\u8f7d\u5e76\u586b\u5199\u7533\u8bf7\u8868\uff08\u7533\u8bf7\u8868\u94fe\u63a5\uff1ahttps:\/\/aka.ms\/InternApplication (opens in new tab)<\/span><\/a> \uff09\u5e76\u5c06\u5176\u4e0e\u5b8c\u6574\u7684\u4e2d\u82f1\u6587\u7b80\u5386\uff08PDF\/Word\u5f62\u5f0f\uff09\u4e00\u540c\u53d1\u9001\u81f3\uff1a MSRAih@microsoft.com<\/a> & lizo@microsoft.com\uff0c\u90ae\u4ef6\u6807\u9898\u4e2d\u6ce8\u660e\uff1a\u673a\u5668\u5b66\u4e60\u7ec4—\u5224\u522b\u5f0f\u57fa\u7840\u6a21\u578b\u4e0e\u9ad8\u6548\u51b3\u7b56\u65b9\u5411\u5b9e\u4e60\u751f<\/strong>\u3002 <\/p>\n\n\n\n

                  \u4e86\u89e3\u66f4\u591a\u201c\u660e\u65e5\u4e4b\u661f\u5b9e\u4e60\u751f\u8ba1\u5212\u201d\uff0c\u8bf7\u8bbf\u95ee\uff1a\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u201c\u660e\u65e5\u4e4b\u661f\u201d\u5b9e\u4e60\u751f\u9879\u76ee – Microsoft Research<\/a> <\/p>\n\n\n\n

                  <\/p>\n","protected":false},"featured_media":0,"parent":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_opportunity_details":[],"msr_opportunity_pubdate":"","msr_hidden_opportunity":false,"msr_opportunity_poc":[],"msr_opportunity_hta":"","msr_opportunity_external_link":"no","msr_opportunity_legal_copy":"","msr-author-ordering":[],"footnotes":""},"research-area":[13556],"msr-region":[197903],"msr-job-opportunity-type":[234821],"msr-city":[243628],"msr-locale":[268875],"msr-program-audience":[],"msr-impact-theme":[],"class_list":["post-1191813","msr-job-opportunity","type-msr-job-opportunity","status-publish","hentry","msr-research-area-artificial-intelligence","msr-region-asia-pacific","msr-job-opportunity-type-internship","msr-city-beijing-china","msr-locale-en_us"],"msr_opportunity_cta":"","msr_opportunity_subdate":"","msr_opportunity_subdate_type":"","msr_opportunity_pubdate":"","msr_opportunity_legal_copy":"","msr_opportunity_poc":[],"msr_opportunity_type":["Internship"],"msr_research_lab":[199560,1012650],"related-researchers":[],"msr_impact_theme":[],"related-groups":[269241],"related-projects":[],"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-job-opportunity\/1191813","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-job-opportunity"}],"about":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-job-opportunity"}],"version-history":[{"count":3,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-job-opportunity\/1191813\/revisions"}],"predecessor-version":[{"id":1191837,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-job-opportunity\/1191813\/revisions\/1191837"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1191813"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1191813"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=1191813"},{"taxonomy":"msr-job-opportunity-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-job-opportunity-type?post=1191813"},{"taxonomy":"msr-city","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-city?post=1191813"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1191813"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=1191813"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1191813"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}