ML Reads

Personal arXiv list

ML papers to read today.

Pick a topic and keep a small daily list of papers worth opening.

Refresh queueDaily mix

Today's queue

5 papers

#01Nov 13, 2026

cs.CV

CoPre-MoE: Consolidating and Preserving MoE for Multimodal Continual Instruction Tuning

Tao Hu, Ling-Xiang Zhang, Da-Wei Zhou

Multimodal continual instruction tuning requires multimodal large language models to sequentially acquire new capabilities without catastrophic forgetting. Mixture-of-experts (MoE) architectures offer a promising way to mitigate task interference through expert specialization and combination. However, while existing MoE-based methods avoid overwriting shared parameters by allocating separate experts for each task, this progressive expansion incurs severe cross-task redundancy and linear parameter growth. Specifically, experts across tasks often learn overlapping functions alongside task-specific capabilities. For instance, ScienceQA and TextVQA both locate relevant image regions. To curb parameter redundancy without inducing forgetting, we propose Consolidating and Preserving MoE (CoPre-MoE). For each incoming task, CoPre-MoE first learns task-specific MoE to capture specialized capabilities. To eliminate redundancy and reduce parameter growth across tasks, we design a shared-and-private expert architecture. Specifically, we first evaluate functional overlap to selectively consolidate overlapping experts across tasks into a fixed-capacity generic expert pool. We then factorize the remaining task-specific experts into shared projection bases and compact task-specific cores within a private expert pool. To mitigate forgetting, we construct the router by pairing each expert with learnable retrieval keys, where historical keys remain frozen during continual updates while newly added keys are regularized with an orthogonality penalty. At inference, input-dependent routing dynamically combines both pools without test-time task labels. Extensive experiments demonstrate that CoPre-MoE achieves superior performance while substantially reducing parameter overhead.

#02Nov 13, 2026

cs.AI

VLAN: Vision-Language Accessible Navigation

Jie Hu, Yu Zheng, Wenzhi Wu and 3 more

Accessible mobility is user-conditioned rather than scene-intrinsic, as walkable paths for one embodiment are inaccessible to another. Yet, existing benchmarks evaluate traversability under a fixed and limited embodiment. We present Vision-Language Accessible Navigation (VLAN), containing 1,000 scenes, 20K navigation trajectories, and 160K keyframes across diverse lighting, environmental conditions, scene semantics, and path geometries. For each trajectory, VLAN provides embodiment-specific actions and accessibility features for legged and wheeled agents, including humanoid and quadruped robots and people with visual or mobility impairments, while everyday obstacles and accessibility labels are categorized as Passable, Bypass, HighRisk, and Blocked. Unlike goal-only evaluation, our benchmark measures physical safety and alignment with user-specific constraints via goal-reaching, collision-aware, and accessibility-aligned metrics. We benchmark modern vision-language models and find they often fail to consistently adapt obstacle relevance, path selection, and safety criteria across user profiles, suggesting accessibility-aware navigation requires dedicated datasets, tasks, and metrics beyond conventional embodied navigation evaluation. All data and code open-source at https://github.com/InSAI-Lab/VLAN.

#03Nov 13, 2026

cs.CV

SWAM: Saliency-Weighted Asymmetric Merging for Efficient Vision Transformers

Utkarsh Kumar Singh, Manish Das, Shiv Ram Dubey and 1 more

Vision Transformers (ViTs) achieve strong performance on image recognition tasks, yet their quadratic self-attention complexity over input tokens creates prohibitive computational costs for practical deployment. Token reduction methods offer a practical solution by progressively reducing the token sequence during the forward pass. We propose Saliency-Weighted Asymmetric Merging (SWAM), a training-based token merging framework that extends the Bipartite Soft Matching (BSM) with two saliency-aware mechanisms. We introduce a two-step matching formulation in which cosine similarity decides candidate pairs, while a multiplicative maximum saliency penalty regulates their selection for merging. We then further use asymmetric exponential saliency-weighted aggregation, which allows more salient tokens to contribute more strongly to the merged representation. A lightweight MLP-based saliency prediction module (75k parameters per stage, 1% of backbone parameters) supervised via attention rollout is trained end-to-end jointly with a DeiT-Small backbone using a composite cross-entropy, knowledge-distillation, and saliency-alignment objective. On CIFAR-100, TinyImageNet, and ImageNet-100, under a conservative schedule that reduces tokens from 196 to 100 patch tokens, with the CLS token retained throughout, 27.5% FLOPs reduction and 1.35 times speedup, SWAM matches or exceeds baseline across the reported results. Under an aggressive schedule (196 to 70 tokens with the CLS token kept throughout, 33.8% FLOPs reduction, and 1.46 times speedup), accuracy gains are achieved on ImageNet-100 (+0.23%) with minimal losses on CIFAR-100 and TinyImageNet.

#04Nov 13, 2026

cs.CV

Deep Bundle Adjustment

Pak Hong Chui, Hector Andrade-Loarca, Daniel Cremers

We present \textbf{DeepBA} (\textbf{Deep} \textbf{B}undle \textbf{A}djustment), a graph neural network (GNN) that recovers camera poses, intrinsics, and scene structure from 2D matches across an arbitrary number of views. Rather than solving a non-linear least-squares problem per scene, DeepBA amortizes bundle adjustment into a learned graph network that iteratively refines its estimate from geometric residuals. Where classical SfM computes its updates analytically and needs careful initialization, DeepBA learns them and converges from a cold start. It is trained with scale-invariant supervision (ground-truth rotation, translation direction, and intrinsics, plus reprojection self-consistency), so datasets with arbitrary metric units can be mixed without per-scene normalization. DeepBA is decoupled from the matcher. At 2.04M parameters, it outperforms prior learned SfM from matches on scene reconstruction and on most pose metrics, is competitive with classical SfM at moderate accuracy thresholds and overtakes it in wall time as the view count grows, and approaches monolithic image-to-geometry transformers such as VGGT ($\sim$1.2B parameters) with greater modularity. Project~page: https://cphjsn.github.io/deep-ba/

#05Nov 13, 2026

cs.GR

TetraCoder: Adaptive Tetrahedral Encoding for Neural Fields

Daniele Della Pietra, Antonella Rech, Nicola Conci and 1 more

Implicit and explicit representations of geometry are central in computer graphics applications and come with their pros and cons. With surface sampling being faster on explicit representations and spatial queries being faster on implicit representations, finding efficient conversions between them is a highly desirable property. We present TetraCoder, a single resolution hash-free adaptive tetrahedral encoding that links an explicit tetrahedral spatial encoder with an underlying implicit signed distance field (SDF). The resulting hybrid implicit-explicit space is continuous piecewise affine (CPWA) by design, allowing to obtain an exact correspondence between the implicit SDF and the explicit analytic mesh. TetraCoder's design comes with additional perks, in particular: (a) it scales better than multi-resolution encoding methods due to its single resolution optimizable grid, (b) unlike hash-based methods, it is not affected by collisions, (c) it enables fast tetrahedral ray-tracing, geometry and appearance editing, directly through its adaptive tetrahedral scaffold. We demonstrate the versatility of TetraCoder by showing an improvement over state-of-the-art methods on multiple tasks: 2D/3D fitting from images and point clouds, 3D novel view synthesis, 3D reconstruction, and geometry/appearance editing.