High schooler's TopoLoss-trained RNNs recover brain-like connectivity
dyamins · x · 2026-10-07
An update from Apurva Ratan's topography work: high schooler Navneet Prakash trained path-integrating RNNs with TopoLoss and found they recover a much more brain-like center-surround receptive-field connectivity structure than baselines, without any post-hoc sorting.
More from Research
- Apple-style compression: LSP learns which subspaces to drop, cutting LLM weights 70% — Massimo Bini · 2026-10-08
- Apple's Stepped MoE: one model scales 1-4B parameters, beating dense counterparts — apple · 2026-10-08
- Gary Marcus: if a system generates a million solutions and one passes Lean, credit the filter? — GaryMarcus · 2026-10-08
- ByteDance Seed Paper Explains Phase Blind Spots in KV Compression Behind DeepSeek's Erratic Long- Context Performance — teortaxesTex · 2026-10-08
- VeriSoftBench: repo-scale Lean 4 verification benchmark, best LLM scores just 41% — xiye_nlp · 2026-10-08
- Perplexity releases pplx-embed-v2-late: OCR-free late-interaction embeddings topping retrieval benchmarks — perplexity_ai · 2026-10-08