arXiv subjects

Kartik Bali

Publications and source records attributed to Kartik Bali.

6 recordsLinked to original sources

arXiv2026

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

Adaptive volumetric finite element meshing traditionally requires iterative PDE solvers or heavily supervised, data-driven surrogates trained on large-scale simulation data. This work explores a significant question: can the high-level semantic understanding of off-the-shelf MLLMs serve as a viable, zero-shot geometric proxy for finite element mesh refinement? We introduce $\textbf{GReFEM}$ (Geometric Reasoning Enhanced Multimodal LLMs for Finite Element Meshing), a framework that utilizes MLLMs to visually localize stress-critical regions based on physics-guided textual prompts. To bridge the gap between 2D MLLM pre-training and 3D geometries, we introduce $\textit{orthoViews}$, a trained view-selection module that assists these MLLMs by maximizing the observability of key geometric features. We evaluate $\textbf{GReFEM}$ by computing finite-element error at stress concentrations, on diverse CAD mesh geometries under multiple loading cases, with MLLMs and prompt variants, against blind and load-informed geometric heuristics, solver-informed Zienkiewicz--Zhu (ZZ) refinement, and a retrained learned adaptive-meshing baseline, all at a matched element budget. Without any FEM solve or simulation training data, $\textbf{GReFEM}$ lowers the critical-region peak von Mises stress error relative to a blind geometric heuristic and to refinement driven by a coarse FEM solve, closing $58\%$ of the gap between the unrefined mesh and a reference-solve oracle, while a hand-derived analytical load-informed heuristic remains stronger and the learned baseline performs slightly better overall. With this we investigate their performance as semantic assistants in automated simulation workflows. Code is available at https://github.com/kbali1297/GReFEM

cs.GR↗

arXiv2026

Look Where You Can: Active View Selection for CAD Reconstruction under Occlusion

CAD reconstruction methods assume a luxury reality rarely grants: unrestricted visual access to the object, photographed from any desired angle. Real objects, however, are scene-embedded, bolted against walls, wedged into corners, resting on floors, where the scene renders much of the view sphere unreachable and the remaining views unequally informative. We introduce \textbf{SightCAD}, a framework for parametric CAD reconstruction that treats view feasibility as a first-class constraint. In this work we consider objects from standard CAD benchmarks embedded in realistic indoor scenes with physically derived visibility constraints over a discrete view sphere. A learned view selector must choose $K$ feasible views for a vision--language model (VLM) that generates executable CadQuery code, scored by geometric fidelity of the executed solid. Because reward arrives only after discrete view selection, autoregressive generation, and CAD-kernel execution, we propose a joint training paradigm in which the view selector and the CAD-generation VLM are trained together against this reward. The learned selection policy departs sharply from random, uniform, and coverage-greedy alternatives, outperforming surface-area maximization (SA-max) by up to $6.4$ Intersection-over-Union (IoU) points across budgets $K\in\{1,\dots,5\}$. The full system surpasses strong external baselines on scene-embedded, occluded multi-view renders of DeepCAD and Fusion360 objects ($+21$ and $+17$ effective-mIoU points over the best baseline, respectively), as well as on test-time domain-canonicalized real images from the industrial T-LESS benchmark and on both synthetic and real images from the MP6D industrial metal-parts benchmark, while producing the highest rate of executable programs of any method compared (invalid-code rate ${\leq}1.5\%$).

cs.CV↗

arXiv2026

Synthetic Persona Pretraining: Alignment from Token Zero

As language-model-based AI is increasingly deployed in autonomous settings, aligning its goals and values with those of humans becomes critical. Today, alignment, and the assistant identity itself, are typically introduced only after pretraining, once behavioral priors are already established. This can make values a thin overlay, rather than deeply rooted, and facilitate subsequent misalignment. Pursuing a different paradigm, we introduce Synthetic Persona Pretraining (SPP), which installs the desired assistant persona from token zero in pretraining. First, we annotate pretraining documents with value-aligned first-person reflections derived from a normative value constitution. Second, we pretrain via the standard cross-entropy loss on standard pretraining documents as well as their reflections, which installs the desired persona among a multitude of other personas. Finally, we post-train on user-assistant dialogue data, which binds this desired persona to the assistant identity, a process we call persona binding. By pretraining models up to 3B parameters on 500B tokens, we show that SPP improves constitution following and jailbreak robustness, and reduces the misalignment rate in out-of-distribution moral dilemmas, while preserving capabilities. Early intervention matters: compared with alignment from token zero, introducing SPP only at the end of pretraining yields weaker constitution adherence, does not shift value priorities, and leads to less aligned choices in dilemmas. This advantage depends on persona binding and, importantly, increases with pretraining budget. Overall, our results show that shaping values early is critical for alignment and establish pretraining-time persona interventions as an effective approach to do so.

cs.LG↗

arXiv2026

MV-GEL: Language-Driven Multi-View Geometric Entity Localization on Meshes

Identifying and grounding precise geometric entities, such as edges, planar regions, and curved surfaces within 3D objects, is foundational to computer-aided design (CAD), robotic manipulation, and scientific simulation. Although modern Vision Language Models (VLMs) have advanced referring segmentation (RIS) in the image domain, extending such language-driven localization to structured 3D geometry is substantially harder. The 3D object appearance is highly sensitive to viewpoints; a single perspective may render a target entity clearly observable, while another may suffer from severe occlusion or foreshortening. In this work, we attempt to solve these challenges with MV-GEL (Multi-View Geometric Entity Localization), a framework for localizing fine-grained geometric entities on polygon meshes from natural language queries. Our key insight is that reliable CAD entity (i.e., faces, edges or solids) localization depends on selecting views that make the queried entity maximally interpretable. We introduce GELviews, a prompt-conditioned ranking module that prioritizes viewpoints based on language prompted observability of geometric CAD entities. Selected views are processed by a VLM-based reasoning segmentation backbone, and predicted masks are lifted to the corresponding meshes via geometry-aware ray casting. Our framework is completely CAD agnostic and relies only on 3D meshes. Experiments show up to a 1.7X improvement in face-level IoU and over 4.5X gains in edge-level F1 compared to vanilla baselines, substantially outperforming CLIP-based and random view sampling, particularly for thin and view-sensitive structures. The dataset, code and trained checkpoints are available at https://github.com/kbali1297/MV-GEL.

cs.CV↗

arXiv2024

Differentiability in Unrolled Training of Neural Physics Simulators on Transient Dynamics

Unrolling training trajectories over time strongly influences the inference accuracy of neural network-augmented physics simulators. We analyze this in three variants of training neural time-steppers. In addition to one-step setups and fully differentiable unrolling, we include a third, less widely used variant: unrolling without temporal gradients. Comparing networks trained with these three modalities disentangles the two dominant effects of unrolling, training distribution shift and long-term gradients. We present detailed study across physical systems, network sizes and architectures, training setups, and test scenarios. It also encompasses two simulation modes: In prediction setups, we rely solely on neural networks to compute a trajectory. In contrast, correction setups include a numerical solver that is supported by a neural network. Spanning these variations, our study provides the empirical basis for our main findings: Non-differentiable but unrolled training with a numerical solver in a correction setup can yield substantial improvements over a fully differentiable prediction setup not utilizing this solver. The accuracy of models trained in a fully differentiable setup differs compared to their non-differentiable counterparts. Differentiable ones perform best in a comparison among correction networks as well as among prediction setups. For both, the accuracy of non-differentiable unrolling comes close. Furthermore, we show that these behaviors are invariant to the physical system, the network architecture and size, and the numerical scheme. These results motivate integrating non-differentiable numerical simulators into training setups even if full differentiability is unavailable. We show the convergence rate of common architectures to be low compared to numerical algorithms. This motivates correction setups combining neural and numerical parts which utilize benefits of both.

physics.comp-ph↗

arXiv2019

Numerical Analysis of Flow Past an Oscillating Circular Cylinder using Arbitrary Lagrangian Eulerian Consistent Flux Reconstruction Scheme

This paper proposes an Arbitrary Lagrangian-Eulerian incompressible finite volume scheme based on Consistent Flux Reconstruction method (ALE-CFR) on deforming two-dimensional unstructured triangular grids. The scheme is further applied to solve for a problem of two-dimensional incompressible viscous flow past a rotationally oscillating circular cylinder when subjected to forced transverse oscillations at Re = 100. The paper first focuses on the mathematical development and validation of the ALE-CFR scheme and then its application on the problem. The rotational oscillation frequencies were taken corresponding to the four primary vortex shedding mode frequencies discovered from the studies conducted by Tokumaru and Dimotakis. The transverse oscillation frequency was varied for a frequency ratio $f_{tv}$ with respect to natural strouhal frequency for flow past a stationary circular cylinder from $0.9 \leq f_{tv} \leq 3.0$. The amplitude ratio for transverse oscillations was taken as 0.2 with respect to the cylinder diameter. The distinct effects on the vortex shedding patterns and the contributions of such rotational and transverse oscillations on lift and drag characteristics of the circular cylinder are studied using the newly developed scheme.

physics.flu-dyn↗