{
 "artifacts": null,
 "category": "geometry",
 "description": "Replace ordinary token pooling or attention aggregation in a hyperbolic representation space with the point satisfying a bounded radial equilibrium law. Each token contributes a unit tangent direction multiplied by \\(\\tanh\\) of its hyperbolic distance from the candidate, so distant outliers cannot dominate the pooled representation while nearby, geometrically consistent tokens still determine it.",
 "formulas_latex": [
  "$$\\sum_i (\\tanh t_i)\\,{\\bf v}_i=0\\qquad\\text{and}\\qquad\\sum_i (\\sinh t_i)\\,{\\bf v}_i=0.$$",
  "$$\\mathbb H^d=\\{x\\in\\mathbb R^{d+1}:\\langle x,x\\rangle_L=-1,\\ x_0\u003e0\\},\\qquad \\langle x,y\\rangle_L=-x_0y_0+\\sum_{k=1}^{d}x_ky_k.$$",
  "$$t_i=d(p,x_i)=\\operatorname{arcosh}(-\\langle p,x_i\\rangle_L),\\qquad u_i=\\frac{\\log_p(x_i)}{t_i},\\qquad \\sum_{i=1}^{m}\\tanh(t_i)u_i=0.$$",
  "$$p_{k+1}=\\operatorname{Exp}_{p_k}\\!\\left(\\eta_k\\,\\frac{1}{m}\\sum_{i=1}^{m}\\tanh(d(p_k,x_i))\\,\\frac{\\log_{p_k}(x_i)}{\\max(d(p_k,x_i),\\varepsilon)}\\right).$$"
 ],
 "id": 2777,
 "implementation": "(1) Integration point: use this operator as a replacement for mean pooling over token embeddings, graph-neighborhood aggregation, or a mixture-of-experts router's summary vector. The encoder outputs points \\(x_i\\) on the hyperboloid; map an unconstrained vector \\(z_i\\in\\mathbb R^d\\) to \\(x_i=(\\sqrt{1+\\|z_i\\|^2},z_i)\\). Use the pooled point \\(p\\) as the sequence representation or routing query.\n\n(2) Pseudocode: initialize \\(p_0\\) as the normalized Lorentz mean or choose \\(p_0=x_1\\). For \\(k=0,\\ldots,K-1\\), compute \\(s_i=-\\langle p_k,x_i\\rangle_L\\), \\(t_i=\\operatorname{arcosh}(\\max(s_i,1+10^{-6}))\\), and \\(\\log_{p_k}(x_i)=\\frac{t_i}{\\sqrt{s_i^2-1}}(x_i-s_i p_k)\\). Set \\(g_k=m^{-1}\\sum_i\\tanh(t_i)\\log_{p_k}(x_i)/\\max(t_i,\\varepsilon)\\), then update \\(p_{k+1}=\\operatorname{Exp}_{p_k}(\\eta_k g_k)\\), where \\(\\operatorname{Exp}_p(v)=\\cosh(\\|v\\|_L)p+\\sinh(\\|v\\|_L)v/\\|v\\|_L\\). Stop after 3-8 iterations or when \\(\\|g_k\\|_L\u003c10^{-4}\\), and backpropagate through the unrolled iterations.\n\n(3) Computed from the paper's mathematics: the distance, unit tangent, bounded \\(\\tanh\\) influence, and equilibrium update are exact. No spectral or distributional quantities need to be estimated. Empirically compare \\(\\tanh(t_i)\\) with \\(t_i\\), \\(\\sinh(t_i)\\), and constant weights; clip \\(\\eta_k\\|g_k\\|_L\\) below 1.0 for numerical stability.\n\n(4) First cheap experiment: train a small hyperbolic graph neural network or hyperbolic transformer on Cora or a synthetic tree-classification benchmark. Compare Euclidean mean pooling, standard hyperbolic Frechet-mean pooling, hyperbolic attention, and this operator using identical encoders and FLOPs. Add controlled corruption by moving 10-40% of token or node embeddings radially outward by factors of 2-10. The predicted signal is nearly unchanged accuracy under distant-minority corruption, lower sensitivity of the pooled point's distance to corrupted tokens, and improved validation accuracy on hierarchical data. Measure iteration overhead and test whether four iterations suffice; success requires robustness at comparable clean-data accuracy and less degradation than \\(\\sinh\\)-weighted aggregation.",
 "math_summary": "The paper's equilibrium law for Julia's zero is \\(\\sum_i (\\tanh t_i)\\mathbf v_i=0\\), where \\(x_i\\) are root points, \\(p\\) is the candidate point, \\(t_i=d(p,x_i)\\) is hyperbolic distance, and \\(\\mathbf v_i\\) is the unit tangent vector at \\(p\\) pointing toward \\(x_i\\). Its contrasting hyperbolic-zero law is \\(\\sum_i(\\sinh t_i)\\mathbf v_i=0\\); because \\(\\tanh t\\leq 1\\) but \\(\\sinh t\\) grows exponentially, the former has bounded per-token influence and the latter amplifies distant points. We adapt the bounded law to token embeddings \\(x_i\\) in the hyperboloid model \\(\\mathbb H^d=\\{x\\in\\mathbb R^{d+1}:\\langle x,x\\rangle_L=-1,x_0\u003e0\\}\\), with Lorentz product \\(\\langle x,y\\rangle_L=-x_0y_0+\\sum_{k=1}^d x_ky_k\\). Define \\(t_i=\\operatorname{arcosh}(-\\langle p,x_i\\rangle_L)\\), \\(u_i=\\log_p(x_i)/t_i\\) for \\(t_i\u003e0\\), and solve \\(\\sum_i\\tanh(t_i)u_i=0\\). The bounded coefficient supplies the robustness mechanism, while the equilibrium equation is equivariant under hyperbolic isometries.",
 "math_tags": [
  "geometry",
  "differential-geometry",
  "dynamical-systems"
 ],
 "ml_areas": [
  "embedding",
  "attention",
  "graph-nn",
  "moe-routing"
 ],
 "paper": {
  "arxiv_id": "2608.27876",
  "arxiv_url": "https://arxiv.org/abs/2608.27876",
  "summary_what_math_gives_to_ml": "The paper identifies two canonical points in the hyperbolic plane through vector equilibrium laws over roots: Julia's zero uses the bounded radial influence \\(\\tanh t\\), while the hyperbolic zero uses the exponentially growing influence \\(\\sinh t\\). The transferable asset is not the binary-form reduction itself, but the distinction between bounded-influence and distance-amplifying geometric aggregation, together with equivariance under hyperbolic isometries. A practical neural adaptation is a hyperbolic pooling or routing operator whose output is the zero of the bounded equilibrium field, giving robustness to a minority of distant or corrupted tokens. This should be tested against Euclidean mean pooling, hyperbolic Frechet means, and hyperbolic attention on hierarchical or adversarially perturbed data.",
  "title": "Equilibrium Laws for Julia's Zero and the Hyperbolic Zero of Binary Forms",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 5,
  "novelty": 6,
  "usefulness": 6
 },
 "solves": [
  "accuracy",
  "generalization",
  "stability"
 ],
 "title": "Bounded-Influence Hyperbolic Pooling",
 "url": "https://synthcore.org/idea/2777/bounded-influence-hyperbolic-pooling",
 "verification": {
  "peer_reviewed": false,
  "status": "unverified",
  "status_label": "Unverified",
  "verdict_source": "deterministic test code (paired-seed permutation statistics)"
 }
}