HER SYMBIOSIS will join Nobel Heroes Day Singapore 2026, from 5–11 October. The event will bring together Nobel laureates, scientists, AI pioneers, innovators and long-term capital to explore the future of science and discovery. #AIforScience# #WomenInScience#

PosteriorBench: going beyond point reconstructions for scientific inverse-problem to full posterior distributions. Many scientific problems involve indirect or partial observations. Multiple physical fields can explain the same measurements. A solver should capture these possibilities, yet reconstruction accuracy alone is misleading. We spent substantial compute to construct high-fidelity reference posteriors across four tasks: Darcy flow inversion, Poisson source recovery, carbon capture and storage, and light transport material inference. These reference data let researchers directly evaluate their solvers using five complementary metrics. One of our key findings: better reconstruction accuracy can coincide with worse posterior recovery. Even strong generative samplers struggle to get both the mean and variance right, often underestimating the uncertainty. Paper: Code: Thanks to Jiachen Yao, Sean Hsu, Xi Deng, and all our coauthors for making this work possible @Caltech #AIforScience# #InverseProblems# #UncertaintyQuantification#

During Geneva Digital Week, HER SYMBIOSIS visited @CERN to explore how AI is transforming particle physics, detector design, and future scientific discovery. Inspired by global collaboration, and by the next generation of women in science. #AIforScience# #CERN# #WomenInSTEM# #AI#

An AI model just rediscovered Kepler's law from latent space, with zero knowledge of orbital mechanics. And the same core architecture spans molecular dynamics, cell prediction, and even generating cancer treatment hypotheses. Title: JEPA-Anything: Learning Predictive Models across Different Worlds URL: In one sentence: it decomposes a target's representation into K orthogonal subspaces, each with its own dedicated predictor — "Orthogonal Predictive Factorization" (OPF) — and applies this single mechanism across radically different domains, from vision and biology to clinical data, control, and molecular dynamics. 🔭 Highlight 1: Rediscovering a physical law from latent space Trained only on orbital motion data, its latent frequency modes recovered Kepler's law f=(2π)⁻¹a⁻³/². The fitted slope was -1.4991 against a theoretical -1.5, with R²=0.9999999. 🧬 Highlight 2: Generating and validating a cancer treatment hypothesis Factor analysis on liver cancer data proposed combining IL-18 and CD73 blockade, which then showed the strongest tumor cell killing in actual patient-derived organoids. ⚛️ Highlight 3: Consistently strong across molecular dynamics and cell prediction It achieved the lowest error across 100-step molecular simulations of water, quartz, paracetamol, and benzene, and also beat prior methods on single-cell perturbation prediction. It's striking that one core architecture spans such wildly different scientific domains this well. #WorldModels# #AIforScience#

🔬 "It feels like having a team of 50 people doing all the work in a day." A Gemini-based multi-agent system generates, debates, and evolves scientific hypotheses, even surfacing a drug candidate that blocks 91% of scarring responses in liver fibrosis. Title: Co-Scientist: A multi-agent AI partner to accelerate research URL: 📝 Overview Co-Scientist is a collaborative multi-agent AI built on Gemini that generates, critiques, and refines novel scientific hypotheses. By automating the hypothesis generation and evaluation cycle, it acts as an AI research partner that accelerates breakthrough discovery. ❓ Challenges Solved Amid information overload and increasingly complex problems, researchers struggle to form breakthrough hypotheses. Connecting scattered facts across vast literature to identify promising research directions is hard. 💡 Methodology & Proposed Approach It organizes specialized agents into three phases. ・Generation phase: a Generation agent proposes novel hypotheses grounded in literature and data, and a Proximity agent clusters them to ensure diverse exploration ・Debate phase: a Reflection agent critiques as a virtual peer reviewer, and a Ranking agent prioritizes via pairwise comparison and Elo-based tournaments ・Evolution phase: an Evolution agent continuously refines and combines top hypotheses, and a Meta-review agent synthesizes final research proposals ・Most of the compute goes to verification, cross-checking claims against ChEMBL, UniProt, web search, and specialized tools like AlphaFold 🎯 Use Cases It is applied across life sciences: antimicrobial resistance, plant immunity, liver fibrosis treatment discovery, ALS mechanisms, cellular aging reversal, infectious-disease protein identification, metabolic disease, and aging biology. 📊 Results ・In liver fibrosis, it identified a drug candidate that blocks 91% of scarring-linked responses ・In cellular aging, it generated genetic leads that rejuvenated cells in the lab and cut screening analysis from months to days ・Over 100 institutions tested it, with collaborators including Stanford, MIT, Cambridge, and Calico ・Enterprise versions are deployed at organizations like Daiichi Sankyo, Bayer Crop Science, and U.S. National Labs #AIforScience# #AIAgents#