Automated Discovery of Improved Constant Weight Binary Codes
2026-03-04T19:59:01Z•352d609c49c6fbfa899a4e57ee45cd66ad62c6531e0afd7b555d9f3775bb0ec0
DatalogLLMMDLOre-polynomials','Reed-Muller','RIS','GNN','beamforming','MIMO'QAOARLIM-LDPSC3-closed-looparxivattention-biascachingcoding-theoryconstant-weight-codesfunction-correcting-codesheuristicinformation-theorylayer-adaptationlocal-differential-privacymodel-compressionmolecular-communicationphysical-layer-securityphysics-informed-MLquantum-federated-learningresearch-collectiontabu-searchwireless-power-control
What happened
Collection of 13 new arXiv submissions (2026-03-03) spanning coding theory, wireless communications, privacy, and machine learning. Highlights include: improved lower bounds for constant-weight binary codes via tabu search and a novel greedy heuristic; repurposing pre-trained Transformers (PC-LLM) for wireless power control with an interference-aware attention bias; integrating local differential privacy into diffusion-based molecular communication (RLIM-LDP); a curvature-weighted, MDL-grounded framework for layer-adaptive LLM capacity allocation and pruning; an information-theoretic study of-
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_math_it
- Record identifier
- 352d609c49c6fbfa899a4e57ee45cd66ad62c6531e0afd7b555d9f3775bb0ec0
- Enrichment time
- 2026-03-04T19:59:01Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.