{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreifumr4hlk2eks5ulth63cykv3a6cirka25vypk3m4f7zvorjhbz2m",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mhmzx3oiv5s2"
  },
  "path": "/t/seeking-arxiv-cs-lg-endorsement-trust-augmented-multi-phase-retrieval-tamr/174480#post_1",
  "publishedAt": "2026-03-22T06:08:17.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "DOI 10.5281/zenodo.18929634",
    "Abstract ID 6359818",
    "Log in to arXiv | arXiv e-print repository"
  ],
  "textContent": "I am Harish Kumar, founder and lead researcher at Quantamix Solutions B.V. (Netherlands). I am seeking an arXiv endorsement for cs.LG to submit my paper on TAMR+ (Trust-Augmented Multi-phase Retrieval).\n\n**Paper:** TAMR+ introduces a three-stage, knowledge graph-enhanced retrieval system with a novel five-dimension trust scoring formula (TRACE) for EU AI Act compliance. Key contributions include deterministic document pre-filtering, multi-phase retrieval (vector ANN + KG alignment + causal density), formula-based scoring with gap attribution, and the EU-RegQA-20 benchmark.\n\n**Already published:**\n\n  * Zenodo: DOI 10.5281/zenodo.18929634\n  * SSRN: Abstract ID 6359818\n  * EPO patent filed: EP26162901.8\n\n\n\n**My endorsement code: 4Z6NZ6**\n\nIf you are a qualified endorser for cs.LG and willing to help, endorsing takes ~2 minutes at Log in to arXiv | arXiv e-print repository . Happy to share the full manuscript.\n\nThank you!\nHarish Kumar | Quantamix Solutions B.V. | harish.kumar@quantamixsolutions.com",
  "title": "Seeking arXiv cs.LG Endorsement — Trust-Augmented Multi-phase Retrieval (TAMR+)"
}