{
  "run_id": "2026-10-10_221032_robot_learning_study",
  "as_of": "2026-10-10T22:10:32+08:00",
  "timezone": "Asia/Shanghai",
  "fetch_started_at": "2026-10-10T22:11:34.356829+08:00",
  "last_request_started_at": "2026-10-10T22:13:40.964921+08:00",
  "fetch_completed_by": "2026-10-10T22:21:25.847172+08:00",
  "fetch_completion_note": "Each request start time was recorded; all requests had completed before receipt_created_at. Exact per-response completion times were not recorded.",
  "receipt_created_at": "2026-10-10T22:21:25.847172+08:00",
  "scope": "12 works / 14 fixed paper versions; targeted reading per source matrix, not all sections. No discovery search or code/model refresh.",
  "records": [
    {
      "name": "sonic",
      "version": "2511.07820v4",
      "url": "https://arxiv.org/html/2511.07820v4",
      "retrieved_at": "2026-10-10T22:11:34.356829+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2511.07820v4",
      "sha256": "55e0841963a0db45dac4994f5f5fa07878e53c5312bd0375d5207654bceca2f9",
      "bytes": 446925,
      "title": "SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control",
      "article_content_verified": true,
      "section_count": 4
    },
    {
      "name": "sonic",
      "version": "2511.07820v4",
      "url": "https://arxiv.org/abs/2511.07820v4",
      "retrieved_at": "2026-10-10T22:11:36.213717+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2511.07820v4",
      "sha256": "4ea1d5be6ad9ce39c1ae4181b09ac572d835ced77031fc1770f7f23ba494c586",
      "bytes": 49646,
      "title": "[2511.07820v4] SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control",
      "submission_history": "Submission history\nFrom: Zhengyi Luo [\nview email\n]\n[v1]\nTue, 11 Nov 2025 04:37:40 UTC (16,929 KB)\n[v2]\nThu, 4 Dec 2025 19:35:21 UTC (16,945 KB)\n[v3]\nThu, 21 May 2026 17:26:49 UTC (27,316 KB)\n[v4]\nThu, 13 Aug 2026 07:13:21 UTC (19,561 KB)\n"
    },
    {
      "name": "playground",
      "version": "2502.08844v1",
      "url": "https://arxiv.org/html/2502.08844v1",
      "retrieved_at": "2026-10-10T22:11:34.357880+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2502.08844v1",
      "sha256": "83cf744bc39629da11bdbd91fee8373f36cf3fdd5ea13ad1e7d8de4ab8c99e5d",
      "bytes": 443908,
      "title": "MuJoCo Playground",
      "article_content_verified": true,
      "section_count": 8
    },
    {
      "name": "playground",
      "version": "2502.08844v1",
      "url": "https://arxiv.org/abs/2502.08844v1",
      "retrieved_at": "2026-10-10T22:11:36.715806+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2502.08844v1",
      "sha256": "d9b95ca9b92c26160bfdaefcebb42a4f8d23c069d454fc08276ac8c8e6257a93",
      "bytes": 40726,
      "title": "[2502.08844v1] MuJoCo Playground",
      "submission_history": "Submission history\nFrom: Kevin Zakka [\nview email\n]\n[v1]\nWed, 12 Feb 2025 23:30:01 UTC (22,625 KB)\n"
    },
    {
      "name": "hilserl",
      "version": "2410.21845v3",
      "url": "https://arxiv.org/html/2410.21845v3",
      "retrieved_at": "2026-10-10T22:11:34.357488+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2410.21845v3",
      "sha256": "86ff6f59b07dd19243177a685d835eca66dcf1190a904c68616094e130152ca8",
      "bytes": 413603,
      "title": "Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning",
      "article_content_verified": true,
      "section_count": 7
    },
    {
      "name": "hilserl",
      "version": "2410.21845v3",
      "url": "https://arxiv.org/abs/2410.21845v3",
      "retrieved_at": "2026-10-10T22:11:36.579681+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2410.21845v3",
      "sha256": "c64e249db417aeae90c1c6644227952ffe362a6eb3aa26247dba270a6ba98960",
      "bytes": 42722,
      "title": "[2410.21845v3] Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning",
      "submission_history": "Submission history\nFrom: Jianlan Luo [\nview email\n]\n[v1]\nTue, 29 Oct 2024 08:12:20 UTC (22,230 KB)\n[v2]\nWed, 6 Nov 2024 03:14:14 UTC (39,202 KB)\n[v3]\nThu, 20 Mar 2025 09:16:05 UTC (40,067 KB)\n"
    },
    {
      "name": "asap",
      "version": "2502.01143v1",
      "url": "https://arxiv.org/html/2502.01143v1",
      "retrieved_at": "2026-10-10T22:11:34.357697+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2502.01143v1",
      "sha256": "a76e3911a0a6f66d30c84f024e0ddfb442b0f8ab1893c314569ed161241f45e3",
      "bytes": 398333,
      "title": "ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills",
      "article_content_verified": true,
      "section_count": 10
    },
    {
      "name": "asap",
      "version": "2502.01143v1",
      "url": "https://arxiv.org/abs/2502.01143v1",
      "retrieved_at": "2026-10-10T22:11:36.395978+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2502.01143v1",
      "sha256": "c717b11998aad21675b19de946d16881179366602905665921aa009e596b61be",
      "bytes": 46389,
      "title": "[2502.01143v1] ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills",
      "submission_history": "Submission history\nFrom: Tairan He [\nview email\n]\n[v1]\nMon, 3 Feb 2025 08:22:46 UTC (29,602 KB)\n[v2]\nSat, 8 Feb 2025 04:12:16 UTC (29,896 KB)\n[v3]\nSat, 26 Apr 2025 03:22:14 UTC (29,897 KB)\n"
    },
    {
      "name": "drive",
      "version": "2610.09943v1",
      "url": "https://arxiv.org/html/2610.09943v1",
      "retrieved_at": "2026-10-10T22:11:37.016192+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2610.09943v1",
      "sha256": "c10c2fe28238e3d39cdd713be2d47c9843196bfa85944732d991fe7986a053e6",
      "bytes": 369243,
      "title": "Many Ways to Succeed: Diversity-Driven RL Fine-Tuning for VLA Generalization",
      "article_content_verified": true,
      "section_count": 6
    },
    {
      "name": "drive",
      "version": "2610.09943v1",
      "url": "https://arxiv.org/abs/2610.09943v1",
      "retrieved_at": "2026-10-10T22:11:38.404189+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2610.09943v1",
      "sha256": "2859ca8e841e4dae969486af879ca23935ef19f5f325471f5a782ee543976f5c",
      "bytes": 42800,
      "title": "[2610.09943v1] Many Ways to Succeed: Diversity-Driven RL Fine-Tuning for VLA Generalization",
      "submission_history": "Submission history\nFrom: Haoru Li [\nview email\n]\n[v1]\nWed, 7 Oct 2026 12:26:21 UTC (9,826 KB)\n"
    },
    {
      "name": "rlinf",
      "version": "2510.06710v3",
      "url": "https://arxiv.org/html/2510.06710v3",
      "retrieved_at": "2026-10-10T22:11:37.146952+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2510.06710v3",
      "sha256": "59cc9fe7c517949e698838dfba2ca4309b6658f144ec84f9701de829bed96c87",
      "bytes": 394346,
      "title": "RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models",
      "article_content_verified": true,
      "section_count": 11
    },
    {
      "name": "rlinf",
      "version": "2510.06710v3",
      "url": "https://arxiv.org/abs/2510.06710v3",
      "retrieved_at": "2026-10-10T22:11:39.239596+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2510.06710v3",
      "sha256": "76b86f93f21ca3584037eeed49392e445a7f7bef62e01d1d721d29eaea2de8a5",
      "bytes": 45210,
      "title": "[2510.06710v3] RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models",
      "submission_history": "Submission history\nFrom: Mingjie Wei [\nview email\n]\n[v1]\nWed, 8 Oct 2025 07:05:13 UTC (4,236 KB)\n[v2]\nSat, 7 Feb 2026 10:46:31 UTC (3,630 KB)\n[v3]\nWed, 12 Aug 2026 08:36:04 UTC (5,242 KB)\n"
    },
    {
      "name": "star",
      "version": "2609.12549v1",
      "url": "https://arxiv.org/html/2609.12549v1",
      "retrieved_at": "2026-10-10T22:11:37.324868+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2609.12549v1",
      "sha256": "c640fe4d8e3eaf9ac9156e23169470c52c0eac2822c4af948ddc760d6507e901",
      "bytes": 183838,
      "title": "STAR: Sparse Tactile Representation Learning in Vision–Tactile–Language–Action Models for Dexterous Manipulation",
      "article_content_verified": true,
      "section_count": 7
    },
    {
      "name": "star",
      "version": "2609.12549v1",
      "url": "https://arxiv.org/abs/2609.12549v1",
      "retrieved_at": "2026-10-10T22:11:38.367571+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2609.12549v1",
      "sha256": "75e076d00e88ce4233ff5e8e1fc3784fbce97d29189e12ffdf77f408229f8995",
      "bytes": 42473,
      "title": "[2609.12549v1] STAR: Sparse Tactile Representation Learning in Vision-Tactile-Language-Action Models for Dexterous Manipulation",
      "submission_history": "Submission history\nFrom: Xiangcheng Liu [\nview email\n]\n[v1]\nFri, 11 Sep 2026 07:57:24 UTC (2,412 KB)\n[v2]\nFri, 18 Sep 2026 16:43:17 UTC (2,412 KB)\n"
    },
    {
      "name": "simpler",
      "version": "2405.05941v1",
      "url": "https://arxiv.org/html/2405.05941v1",
      "retrieved_at": "2026-10-10T22:11:37.464617+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2405.05941v1",
      "sha256": "f67311ee843383892736d9d51d2ab403f7aee661d74fcb8ec6c29a211da0e25b",
      "bytes": 477186,
      "title": "Evaluating Real-World Robot Manipulation Policies in Simulation",
      "article_content_verified": true,
      "section_count": 8
    },
    {
      "name": "simpler",
      "version": "2405.05941v1",
      "url": "https://arxiv.org/abs/2405.05941v1",
      "retrieved_at": "2026-10-10T22:11:38.779639+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2405.05941v1",
      "sha256": "86ac0bc7b78740ca74e9f8dc64472aa54f3dd3ff7c6039ba4953a86ddde7cafc",
      "bytes": 43590,
      "title": "[2405.05941v1] Evaluating Real-World Robot Manipulation Policies in Simulation",
      "submission_history": "Submission history\nFrom: Xuanlin Li [\nview email\n]\n[v1]\nThu, 9 May 2024 17:30:16 UTC (8,759 KB)\n"
    },
    {
      "name": "robotwin",
      "version": "2506.18088v2",
      "url": "https://arxiv.org/html/2506.18088v2",
      "retrieved_at": "2026-10-10T22:11:39.189287+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2506.18088v2",
      "sha256": "ccbb13031686d408b97a3526e565cdea95fe72ce7e590d08abeb7fe35fc66570",
      "bytes": 1022061,
      "title": "RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation",
      "article_content_verified": true,
      "section_count": 7
    },
    {
      "name": "robotwin",
      "version": "2506.18088v2",
      "url": "https://arxiv.org/abs/2506.18088v2",
      "retrieved_at": "2026-10-10T22:11:45.849700+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2506.18088v2",
      "sha256": "8a10c2354da0a9e155a570efbfc63babf5efd2fb26a9c08bd3b4d968b0e34be9",
      "bytes": 48478,
      "title": "[2506.18088v2] RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation",
      "submission_history": "Submission history\nFrom: Tianxing Chen [\nview email\n]\n[v1]\nSun, 22 Jun 2025 16:26:53 UTC (6,813 KB)\n[v2]\nWed, 27 Aug 2025 17:52:42 UTC (6,909 KB)\n"
    },
    {
      "name": "dreamtrue",
      "version": "2610.12468v1",
      "url": "https://arxiv.org/html/2610.12468v1",
      "retrieved_at": "2026-10-10T22:11:44.385074+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2610.12468v1",
      "sha256": "57cc54c90536c24609fa68950ac00fcb2764a2712357b4b99f5389ed95c3337b",
      "bytes": 374433,
      "title": "DreamTrue: Action-Faithful Robot World Model with Counterfactual Post-Training",
      "article_content_verified": true,
      "section_count": 5
    },
    {
      "name": "dreamtrue",
      "version": "2610.12468v1",
      "url": "https://arxiv.org/abs/2610.12468v1",
      "retrieved_at": "2026-10-10T22:11:46.033171+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2610.12468v1",
      "sha256": "16deefa52fa042d0a104b6d5e010119453278daf8558f1ebc0452bbb88bdafc0",
      "bytes": 43354,
      "title": "[2610.12468v1] DreamTrue: Action-Faithful Robot World Model with Counterfactual Post-Training",
      "submission_history": "Submission history\nFrom: Lue Fan [\nview email\n]\n[v1]\nThu, 8 Oct 2026 17:59:51 UTC (25,713 KB)\n"
    },
    {
      "name": "recap",
      "version": "2511.14759v2",
      "url": "https://arxiv.org/html/2511.14759v2",
      "retrieved_at": "2026-10-10T22:11:44.484065+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2511.14759v2",
      "sha256": "ff52f6a18bb2e191870bea58450a6abd448e5060b29fde4c1e78ab178b3111f6",
      "bytes": 400932,
      "title": "𝜋^∗_0.6: a VLA That Learns From Experience",
      "article_content_verified": true,
      "section_count": 8
    },
    {
      "name": "recap",
      "version": "2511.14759v2",
      "url": "https://arxiv.org/abs/2511.14759v2",
      "retrieved_at": "2026-10-10T22:11:46.296383+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2511.14759v2",
      "sha256": "986011000c3ac87f902c6ab4f5e7361b25581530d8cd460cea11f66e5c36c5c5",
      "bytes": 50105,
      "title": "[2511.14759v2] $π^{*}_{0.6}$: a VLA That Learns From Experience",
      "submission_history": "Submission history\nFrom: Brian Ichter [\nview email\n]\n[v1]\nTue, 18 Nov 2025 18:58:55 UTC (22,750 KB)\n[v2]\nWed, 19 Nov 2025 04:34:49 UTC (22,750 KB)\n"
    },
    {
      "name": "rapolicy",
      "version": "2609.22888v1",
      "url": "https://arxiv.org/html/2609.22888v1",
      "retrieved_at": "2026-10-10T22:11:45.789297+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2609.22888v1",
      "sha256": "fb056ed71861d56e9a3c88689d9053e361d7df7717641b80a661dab483005c86",
      "bytes": 180533,
      "title": "Stable and Efficient Real-World Online VLA Post-Training viaAsynchronous Replay-Anchored Policy Improvement",
      "article_content_verified": true,
      "section_count": 5
    },
    {
      "name": "rapolicy",
      "version": "2609.22888v1",
      "url": "https://arxiv.org/abs/2609.22888v1",
      "retrieved_at": "2026-10-10T22:11:47.131330+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2609.22888v1",
      "sha256": "873e89a2057b21b7bd2c8bc5a7bdb8c9c6d9d435e79b1e490f4f21f4605e5dfb",
      "bytes": 42936,
      "title": "[2609.22888v1] Stable and Efficient Real-World Online VLA Post-Training via Asynchronous Replay-Anchored Policy Improvement",
      "submission_history": "Submission history\nFrom: Jiarui Yang [\nview email\n]\n[v1]\nSat, 19 Sep 2026 08:45:18 UTC (2,723 KB)\n"
    },
    {
      "name": "asap_v3",
      "version": "2502.01143v3",
      "url": "https://arxiv.org/html/2502.01143v3",
      "retrieved_at": "2026-10-10T22:13:09.306475+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2502.01143v3",
      "sha256": "9376d64729ec59f3697c93a9e473abe5b84ac5a14242e933e2ede4ca1f52b8a9",
      "bytes": 398368,
      "title": "ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills",
      "article_content_verified": true,
      "section_count": 10
    },
    {
      "name": "asap_v3",
      "version": "2502.01143v3",
      "url": "https://arxiv.org/abs/2502.01143v3",
      "retrieved_at": "2026-10-10T22:13:10.916888+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2502.01143v3",
      "sha256": "d35ed269c62bd2b3d82ac3b35cdd09f219ee0182fd5327792b3b1e9fa4138b54",
      "bytes": 46392,
      "title": "[2502.01143v3] ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills",
      "submission_history": "Submission history\nFrom: Tairan He [\nview email\n]\n[v1]\nMon, 3 Feb 2025 08:22:46 UTC (29,602 KB)\n[v2]\nSat, 8 Feb 2025 04:12:16 UTC (29,896 KB)\n[v3]\nSat, 26 Apr 2025 03:22:14 UTC (29,897 KB)\n"
    },
    {
      "name": "star_v2",
      "version": "2609.12549v2",
      "url": "https://arxiv.org/html/2609.12549v2",
      "retrieved_at": "2026-10-10T22:13:40.023130+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/html/2609.12549v2",
      "sha256": "1e2146a18715b49fc0df8c20c09ce6778a7484de027ca9495d32fd14041b2dfd",
      "bytes": 184918,
      "title": "STAR: Sparse Tactile Representation Learning in Vision–Tactile–Language–Action Models for Dexterous Manipulation",
      "article_content_verified": true,
      "section_count": 7
    },
    {
      "name": "star_v2",
      "version": "2609.12549v2",
      "url": "https://arxiv.org/abs/2609.12549v2",
      "retrieved_at": "2026-10-10T22:13:40.964921+08:00",
      "status": 200,
      "final_url": "https://arxiv.org/abs/2609.12549v2",
      "sha256": "c5e2bf5f035fd47529f5879419a82a2ce1d233627db59a9db1c54e60c601b02f",
      "bytes": 42280,
      "title": "[2609.12549v2] STAR: Sparse Tactile Representation Learning in Vision-Tactile-Language-Action Models for Dexterous Manipulation",
      "submission_history": "Submission history\nFrom: Xiangcheng Liu [\nview email\n]\n[v1]\nFri, 11 Sep 2026 07:57:24 UTC (2,412 KB)\n[v2]\nFri, 18 Sep 2026 16:43:17 UTC (2,412 KB)\n"
    }
  ]
}