Learning Pseudo-Riemannian Manifolds for Heterophilic Graphs via Graph Signature

Yun Young Choi, Asung Kil, Sun Woo Park, Minho Lee and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Can Folding Models Tell Binders from Bluffers? Evidence from POISK: The Patent-Derived Antibody Dataset

Daria Tupikina, Andrea Roncoli, Alexander Bujotzek, Brennan Abanades Kenyon

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

RoSA: Rotational Sparse Adaptation for Memory-Efficient Fine-Tuning

Muhammad Azeem Lodhi, chao zhou, Rebekka Burkholz

Paris Poster Session 4, Thu, Dec 10, 5:30 PM–7:30 PM, Paris Poster Hall · Published 2026

Understanding the Curse of Unrolling

Non-asymptotic analysis explains the curse of unrolling, early derivative divergence when differentiating through iterative algorithms, and shows that truncating early iterations mitigates it while reducing memory, with warm-starting providing implicit truncation in bilevel optimization.

Sheheryar Mehmood, Florian Knoll, Peter Ochs

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

GReFEM: Multimodal LLMs as Zero-Shot Semantic Assistants for Physics-Guided 3D Mesh Refinement

GReFEM uses multimodal LLMs as zero-shot semantic assistants to localize stress-critical 3D regions and refine finite element meshes more precisely than geometric heuristics.

Kartik Bali, Mahish Kumar Guru, Christian J Cyron, Roland Aydin

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Co-PiLOT: Constrained Physics-Informed Latent Optimization for Target-Driven Inverse Design

Co-PiLOT combines generative latent optimization with physics-informed black-box search to inverse-design magnesium alloy microstructures, cutting relative target error by 3, 22% over seven baselines within 160 simulations.

Mahish Kumar Guru, Mayank Nagar, Ayush vyas, Jan Bohlen and 2 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Semantic Motion Anchors: Bridging Motion and Meaning in Co-Speech Gestures

Semantic motion anchors discretize gesture motion into verbalized primitives to align text and gestures, improving retrieval and generation by capturing communicative intent over low-level kinematics.

Varsha Suresh, Mohammad Mahdi Abootorabi, Mohamed Salman, M. Hamza Mughal and 4 more

Paris Poster Session 4, Thu, Dec 10, 5:30 PM–7:30 PM, Paris Poster Hall · Published 2026