{
  "updated": "2026-09-04T22:27:24.062Z",
  "generatedBy": "scripts/update-data.mjs",
  "arxiv": {
    "totalKD": 5372,
    "last30d": 78,
    "perYear": [
      {
        "year": 2015,
        "count": 2
      },
      {
        "year": 2016,
        "count": 6
      },
      {
        "year": 2017,
        "count": 20
      },
      {
        "year": 2018,
        "count": 54
      },
      {
        "year": 2019,
        "count": 163
      },
      {
        "year": 2020,
        "count": 342
      },
      {
        "year": 2021,
        "count": 467
      },
      {
        "year": 2022,
        "count": 641
      },
      {
        "year": 2023,
        "count": 807
      },
      {
        "year": 2024,
        "count": 1016
      },
      {
        "year": 2025,
        "count": 1122
      },
      {
        "year": 2026,
        "count": 731
      }
    ],
    "recent": [
      {
        "title": "Computationally Efficient Pathology Segmentation using Knowledge Distillation from Foundation Models",
        "url": "http://arxiv.org/abs/2609.03947v1",
        "published": "2026-09-03T14:52:00Z",
        "authors": [
          "Jiaqi Lv",
          "Yijie Zhu",
          "Saki Okada"
        ]
      },
      {
        "title": "Closing the Semantic-Edge Gap: Tiny Language Models for 6G Wireless Intelligence",
        "url": "http://arxiv.org/abs/2609.03747v1",
        "published": "2026-09-03T12:18:09Z",
        "authors": [
          "Srikanth Kamath",
          "Arnav Mathur",
          "Joslyn Sajan George"
        ]
      },
      {
        "title": "ALRA: Adaptive Local Relational Alignment for Logit-Based Pre-training Distillation of Autoregressive Language Models",
        "url": "http://arxiv.org/abs/2609.03355v1",
        "published": "2026-09-03T04:26:39Z",
        "authors": [
          "Quang Hoang Trung",
          "Quang Huu Hieu",
          "Nguyen Van Hoang Phuc"
        ]
      },
      {
        "title": "ProgResViT: Progressive Resolution and Width for Adaptive Vision Transformers",
        "url": "http://arxiv.org/abs/2609.03216v1",
        "published": "2026-09-02T23:10:56Z",
        "authors": [
          "Ali Hojjat",
          "Janek Haberer",
          "Olaf Landsiedel"
        ]
      },
      {
        "title": "Convergence Theory of Knowledge Distillation in Asynchronous P2P Gossip Learning Network",
        "url": "http://arxiv.org/abs/2609.01952v1",
        "published": "2026-09-01T23:52:01Z",
        "authors": [
          "Lucas Qingyang Fang",
          "Tiyao Liu",
          "Jinhao Jing"
        ]
      },
      {
        "title": "Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall",
        "url": "http://arxiv.org/abs/2609.01532v1",
        "published": "2026-09-01T17:00:30Z",
        "authors": [
          "Jacqueline He",
          "Howard Yen",
          "Shuyue Stella Li"
        ]
      },
      {
        "title": "Position Matters: Feature Inversion Attacks in ViT Split Inference with Token Reduction and Shuffling",
        "url": "http://arxiv.org/abs/2609.01232v1",
        "published": "2026-09-01T13:32:30Z",
        "authors": [
          "Stefano Leggio",
          "Giulio Rossolini",
          "Alessandro Biondi"
        ]
      },
      {
        "title": "Freshness-Aware Constrained Sensing-Aided Beam Prediction with Knowledge Distillation",
        "url": "http://arxiv.org/abs/2609.01225v1",
        "published": "2026-09-01T13:27:03Z",
        "authors": [
          "Abolfazl Zakeri",
          "Nhan Thanh Nguyen",
          "Ahmed Alkhateeb"
        ]
      },
      {
        "title": "A Frequency-Aware Dynamic Knowledge Distillation Framework: An Effective Tool for Bridging Low- and High-Frequency Seismic Information",
        "url": "http://arxiv.org/abs/2609.00828v1",
        "published": "2026-09-01T07:30:27Z",
        "authors": [
          "Jun Ma",
          "Xinyang Wang",
          "Xintong Dong"
        ]
      },
      {
        "title": "A Closed-Loop Evaluation of Capability Loss and Recovery in Compressed Driving Policies",
        "url": "http://arxiv.org/abs/2609.00718v1",
        "published": "2026-09-01T04:50:17Z",
        "authors": [
          "Ahmad Alfan Alfian Irfan",
          "Nur Ahmad Khatim",
          "Mansur Arief"
        ]
      },
      {
        "title": "CRAD: Class-wise Reliability-Aware Distillation for Decentralized Heterogeneous Federated Learning",
        "url": "http://arxiv.org/abs/2609.00446v1",
        "published": "2026-08-31T22:36:30Z",
        "authors": [
          "Baraa Bilbeisi",
          "Mengchen Fan",
          "Baocheng Geng"
        ]
      },
      {
        "title": "Uncertainty-Aware Trajectory Forecasting from Imperfect Tracking",
        "url": "http://arxiv.org/abs/2608.30899v1",
        "published": "2026-08-31T14:50:38Z",
        "authors": [
          "Stephane Da Silva Martins",
          "Victor Petrovic",
          "Emanuel Aldea"
        ]
      }
    ]
  },
  "huggingface": {
    "distillModels": 16329,
    "top": [
      {
        "id": "facebook/nllb-200-distilled-600M",
        "downloads": 1458914,
        "likes": 966,
        "url": "https://huggingface.co/facebook/nllb-200-distilled-600M"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
        "downloads": 562818,
        "likes": 1608,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B"
      },
      {
        "id": "distil-whisper/distil-large-v3",
        "downloads": 542836,
        "likes": 379,
        "url": "https://huggingface.co/distil-whisper/distil-large-v3"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
        "downloads": 461220,
        "likes": 1571,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B"
      },
      {
        "id": "empero-ai/Qwen3.8-4B-Distill-GGUF",
        "downloads": 445387,
        "likes": 96,
        "url": "https://huggingface.co/empero-ai/Qwen3.8-4B-Distill-GGUF"
      },
      {
        "id": "empero-ai/Qwen3.8-2B-Distill-GGUF",
        "downloads": 444462,
        "likes": 125,
        "url": "https://huggingface.co/empero-ai/Qwen3.8-2B-Distill-GGUF"
      },
      {
        "id": "empero-ai/Qwen3.8-9B-Distill-GGUF",
        "downloads": 409204,
        "likes": 208,
        "url": "https://huggingface.co/empero-ai/Qwen3.8-9B-Distill-GGUF"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
        "downloads": 396404,
        "likes": 676,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B"
      },
      {
        "id": "Jackrong/Qwen3.5-4B-Claude-4.6-Opus-Reasoning-Distilled-GGUF",
        "downloads": 389107,
        "likes": 158,
        "url": "https://huggingface.co/Jackrong/Qwen3.5-4B-Claude-4.6-Opus-Reasoning-Distilled-GGUF"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
        "downloads": 379197,
        "likes": 875,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
      },
      {
        "id": "facebook/nllb-200-distilled-1.3B",
        "downloads": 377705,
        "likes": 185,
        "url": "https://huggingface.co/facebook/nllb-200-distilled-1.3B"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
        "downloads": 367771,
        "likes": 881,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B"
      },
      {
        "id": "hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-GGUF",
        "downloads": 319789,
        "likes": 292,
        "url": "https://huggingface.co/hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-GGUF"
      },
      {
        "id": "Jackrong/Qwen3.5-9B-GLM5.1-Distill-v1-GGUF",
        "downloads": 316844,
        "likes": 112,
        "url": "https://huggingface.co/Jackrong/Qwen3.5-9B-GLM5.1-Distill-v1-GGUF"
      },
      {
        "id": "mradermacher/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-i1-GGUF",
        "downloads": 309303,
        "likes": 122,
        "url": "https://huggingface.co/mradermacher/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-i1-GGUF"
      }
    ],
    "tracked": [
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
        "downloads": 562818,
        "likes": 1608,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
        "downloads": 367771,
        "likes": 881,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
        "downloads": 379197,
        "likes": 875,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
      },
      {
        "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
        "downloads": 461220,
        "likes": 1571,
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B"
      },
      {
        "id": "distilbert/distilbert-base-uncased",
        "downloads": 7067963,
        "likes": 1127,
        "url": "https://huggingface.co/distilbert/distilbert-base-uncased"
      },
      {
        "id": "Qwen/Qwen3-8B",
        "downloads": 13232997,
        "likes": 1343,
        "url": "https://huggingface.co/Qwen/Qwen3-8B"
      },
      {
        "id": "google/gemma-3-4b-it",
        "downloads": 1626709,
        "likes": 1470,
        "url": "https://huggingface.co/google/gemma-3-4b-it"
      },
      {
        "id": "meta-llama/Llama-3.2-3B-Instruct",
        "downloads": 1419885,
        "likes": 2512,
        "url": "https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct"
      },
      {
        "id": "microsoft/Phi-4-mini-instruct",
        "downloads": 456948,
        "likes": 829,
        "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct"
      },
      {
        "id": "HuggingFaceTB/SmolLM3-3B",
        "downloads": 581340,
        "likes": 1022,
        "url": "https://huggingface.co/HuggingFaceTB/SmolLM3-3B"
      }
    ]
  },
  "github": {
    "repos": [
      {
        "repo": "huggingface/trl",
        "stars": 19223,
        "forks": 2958,
        "pushedAt": "2026-09-04T21:53:29Z",
        "url": "https://github.com/huggingface/trl"
      },
      {
        "repo": "arcee-ai/DistillKit",
        "stars": 1052,
        "forks": 141,
        "pushedAt": "2026-05-12T20:10:14Z",
        "url": "https://github.com/arcee-ai/DistillKit"
      },
      {
        "repo": "pytorch/torchtune",
        "stars": 5804,
        "forks": 752,
        "pushedAt": "2026-09-04T11:38:16Z",
        "url": "https://github.com/meta-pytorch/torchtune"
      },
      {
        "repo": "NVIDIA/NeMo",
        "stars": 18388,
        "forks": 3603,
        "pushedAt": "2026-09-04T21:07:43Z",
        "url": "https://github.com/NVIDIA-NeMo/Speech"
      },
      {
        "repo": "axolotl-ai-cloud/axolotl",
        "stars": 12440,
        "forks": 1421,
        "pushedAt": "2026-09-04T13:14:58Z",
        "url": "https://github.com/axolotl-ai-cloud/axolotl"
      },
      {
        "repo": "unslothai/unsloth",
        "stars": 75624,
        "forks": 6872,
        "pushedAt": "2026-09-04T22:22:29Z",
        "url": "https://github.com/unslothai/unsloth"
      },
      {
        "repo": "hiyouga/LLaMA-Factory",
        "stars": 74577,
        "forks": 9141,
        "pushedAt": "2026-09-04T08:15:24Z",
        "url": "https://github.com/hiyouga/LlamaFactory"
      },
      {
        "repo": "vllm-project/vllm",
        "stars": 90979,
        "forks": 21712,
        "pushedAt": "2026-09-04T22:13:58Z",
        "url": "https://github.com/vllm-project/vllm"
      },
      {
        "repo": "sgl-project/sglang",
        "stars": 35476,
        "forks": 8557,
        "pushedAt": "2026-09-04T22:22:31Z",
        "url": "https://github.com/sgl-project/sglang"
      },
      {
        "repo": "huggingface/open-r1",
        "stars": 26452,
        "forks": 2448,
        "pushedAt": "2026-04-02T14:03:15Z",
        "url": "https://github.com/huggingface/open-r1"
      },
      {
        "repo": "open-thoughts/open-thoughts",
        "stars": 2331,
        "forks": 194,
        "pushedAt": "2025-12-02T21:57:07Z",
        "url": "https://github.com/open-thoughts/open-thoughts"
      },
      {
        "repo": "EleutherAI/lm-evaluation-harness",
        "stars": 13890,
        "forks": 3538,
        "pushedAt": "2026-09-01T13:51:29Z",
        "url": "https://github.com/EleutherAI/lm-evaluation-harness"
      },
      {
        "repo": "SafeAILab/EAGLE",
        "stars": 2524,
        "forks": 295,
        "pushedAt": "2026-02-20T02:53:29Z",
        "url": "https://github.com/SafeAILab/EAGLE"
      },
      {
        "repo": "NovaSky-AI/SkyThought",
        "stars": 3399,
        "forks": 344,
        "pushedAt": "2025-07-12T03:05:27Z",
        "url": "https://github.com/NovaSky-AI/SkyThought"
      },
      {
        "repo": "simplescaling/s1",
        "stars": 6667,
        "forks": 756,
        "pushedAt": "2025-06-25T18:59:18Z",
        "url": "https://github.com/simplescaling/s1"
      }
    ]
  },
  "news": {
    "items": [
      {
        "title": "Show HN: Distilling DeepSeek into GPT-OSS doesn't transfer censorship. Try it",
        "url": "https://www.ctgt.ai/research/distillation-censorship-transfer",
        "source": "ctgt.ai",
        "points": 170,
        "date": "2026-07-30T18:13:06Z"
      },
      {
        "title": "The tech world is suddenly obsessed with one concept in AI: Distillation",
        "url": "https://www.cnbc.com/2026/07/25/hat-is-distillation-and-why-is-everyone-so-obsessed-with-it-this-week.html",
        "source": "cnbc.com",
        "points": 3,
        "date": "2026-07-25T14:58:39Z"
      },
      {
        "title": "Top White House official escalating the fight over Moonshot AI's Kimi K3 model",
        "url": "https://www.businessinsider.com/white-house-kimi-k3-moonshot-ai-distillation-2026-7",
        "source": "businessinsider.com",
        "points": 7,
        "date": "2026-07-22T19:02:59Z"
      },
      {
        "title": "Updating IP Regulations for AI Distillation",
        "url": "https://www.marble.onl/posts/copyright_vs_ai.html",
        "source": "marble.onl",
        "points": 2,
        "date": "2026-07-19T00:14:11Z"
      },
      {
        "title": "Responding to AI Distillation Without Panic",
        "url": "https://www.lawfaremedia.org/article/responding-to-ai-distillation-without-panic",
        "source": "lawfaremedia.org",
        "points": 2,
        "date": "2026-07-16T21:08:25Z"
      },
      {
        "title": "A brief history of distillation in AI",
        "url": "https://twitter.com/SergioPaniego/status/2073066275819991472",
        "source": "twitter.com",
        "points": 2,
        "date": "2026-07-13T18:50:32Z"
      },
      {
        "title": "American A.I. Companies Say Chinese Copycats Are Quickly Catching Up",
        "url": "https://www.nytimes.com/2026/07/06/technology/ai-distillation-china.html",
        "source": "nytimes.com",
        "points": 8,
        "date": "2026-07-06T17:23:03Z"
      },
      {
        "title": "Anthropic Accuses Alibaba of Largest AI Distillation Attack: 28.8M Fraudulent",
        "url": "https://yipzap.com/anthropic-accuses-alibaba-of-largest-ai-distillation-attack-28-8m-fraudulent-exchanges/",
        "source": "yipzap.com",
        "points": 4,
        "date": "2026-06-26T00:49:44Z"
      },
      {
        "title": "96% Correct Next Token Prediction, with No DNN, No Training, Autodistilled Model",
        "url": "https://mltechniques.com/2026/05/25/96-correct-next-token-prediction-with-no-dnn-no-training-auto-distilled-model/",
        "source": "mltechniques.com",
        "points": 3,
        "date": "2026-05-25T06:21:31Z"
      },
      {
        "title": "I over-engineered my simple AI backend: distillation, router, embedding etc.",
        "url": "https://sisyphusconsulting.org/case-studies/2026/04/01/scaling-llms-at-the-edge/",
        "source": "sisyphusconsulting.org",
        "points": 2,
        "date": "2026-04-30T14:25:53Z"
      },
      {
        "title": "US accuses China of industrial-scale AI model distillation, will share Intel",
        "url": "https://thenextweb.com/news/us-white-house-ai-model-distillation-china-theft",
        "source": "thenextweb.com",
        "points": 11,
        "date": "2026-04-29T21:16:35Z"
      },
      {
        "title": "When model distillation becomes a diplomatic incident",
        "url": "https://underlines.news/2026/04/26/us-orders-global-diplomatic-warning-on-chinese-distillation-of-ai-models",
        "source": "underlines.news",
        "points": 1,
        "date": "2026-04-28T09:22:28Z"
      },
      {
        "title": "White House Memo on Adversarial Distillation of American AI Models [pdf]",
        "url": "https://whitehouse.gov/wp-content/uploads/2026/04/NSTM-4.pdf",
        "source": "whitehouse.gov",
        "points": 7,
        "date": "2026-04-25T00:53:13Z"
      },
      {
        "title": "When Distillation Strips the Soul: Safety Comparison of a Claude-Distilled Model",
        "url": "https://netrork.com/blog/when-distillation-strips-the-soul/",
        "source": "netrork.com",
        "points": 1,
        "date": "2026-03-08T01:46:51Z"
      },
      {
        "title": "Show HN: Aside – Local meeting capture with vault-native AI distillation",
        "url": "https://github.com/jshph/aside/",
        "source": "github.com",
        "points": 1,
        "date": "2026-03-03T15:27:35Z"
      },
      {
        "title": "AI could eat itself: Competitors (..) steal their secrets and clone them",
        "url": "https://www.theregister.com/2026/02/14/ai_risk_distillation_attacks/",
        "source": "theregister.com",
        "points": 12,
        "date": "2026-02-14T11:59:12Z"
      },
      {
        "title": "Google identifies over 100k prompts used in distillation attacks",
        "url": "https://cloud.google.com/blog/topics/threat-intelligence/distillation-experimentation-integration-ai-adversarial-use",
        "source": "cloud.google.com",
        "points": 7,
        "date": "2026-02-12T21:16:33Z"
      },
      {
        "title": "Show HN: Feedstack – SerpAPI/OpenAI extract feature requests from Reddit/forums",
        "url": "https://feedstack.app",
        "source": "feedstack.app",
        "points": 2,
        "date": "2026-01-07T16:07:57Z"
      },
      {
        "title": "Show HN: Reverse Jailbreaking a Psychopathic AI via Identity Injection",
        "url": "https://github.com/DRawson5570/AI-Wisdom-Distillation",
        "source": "github.com",
        "points": 5,
        "date": "2025-11-22T20:33:51Z"
      },
      {
        "title": "Distillation makes AI models smaller and cheaper",
        "url": "https://www.quantamagazine.org/how-distillation-makes-ai-models-smaller-and-cheaper-20250718/",
        "source": "quantamagazine.org",
        "points": 164,
        "date": "2025-07-20T22:09:27Z"
      }
    ]
  },
  "errors": []
}
