{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreie7v2qmtkiyje7fkaojzsc7jk5qkzjhussvcape5bdqjhhc7oovmi",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mlubbhs7pqf2"
  },
  "path": "/t/title-arxiv-endorsement-request-from-weight-space-diffusion-to-latent-space-deepsdf-cs-cv-cs-gr/176028#post_1",
  "publishedAt": "2026-05-15T01:06:48.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "https://jainaditya.in/whitepaper/hypernet-deepsdf"
  ],
  "textContent": "Hello community,\n\nI am seeking an arXiv endorsement for a recent research project investigating image-to-3D generation. Due to the updated 2026 submission policies, I am looking for a peer review and endorsement from the community.\n\nPaper Title: From Weight-Space Diffusion to Latent-Space DeepSDF: An Empirical Investigation of Image-Conditioned 3-D Generation at Small Data Scale\n\nTarget Categories: * Primary: Computer Vision and Pattern Recognition (cs.CV)\n\n  * Secondary: Graphics (cs.GR); Machine Learning (cs.LG)\n\n\n\nAbstract: We present a systematic investigation of 3D shape generation under severe data constraints (≤976 shapes). The work traces the structural failure modes of 54,785-dimensional weight-space diffusion—specifically the “warm-start dominance” problem (0.96 mean cosine similarity)—and demonstrates a successful pivot to a DINOv2-conditioned Latent Diffusion Model (LDM) using DeepSDF embeddings. Our results show that architectural inductive biases are more critical for out-of-distribution generalization than learned compression at this data scale.\n\nLink to PDF: https://jainaditya.in/whitepaper/hypernet-deepsdf\n\nResearcher Details: * Author: Aaditya Jain\n\n  * ORCID: 0009-0005-5534-5641\n\n  * Affiliation: Independent Thesis Research\n\n\n\n\nI have documented the twelve experimental phases, including the failure of weight-space autoencoders and the success of the latent DiT pipeline. I am happy to provide my arXiv ID or discuss the technical logs (Phase 1-12) with anyone willing to review the work for endorsement.\n\nThank you for supporting independent research in 3D machine learning!",
  "title": "Title: arXiv Endorsement Request: From Weight-Space Diffusion to Latent-Space DeepSDF - [cs.CV / cs.GR]"
}