2026-08-07 AI 리서치 브리핑

ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Deaf and hard-of-hearing people in Bangladesh communicate mainly through Bangla Sign Language (BdSL). Automatic BdSL recognition on personal devices could widen access to education and services. Existing systems use controlled-setting datasets without expert verification and heavyweight pretrained backbones unsuited to on-device use. We introduce RSBdSL38, 10,874 expert-validated images spanning all 38 BdSL hand signs, representing the 51 letters of the Bangla alphabet, recorded from real signers at three special-needs schools across Bangladesh. We propose a lightweight attention based convolutional network of 298,470 parameters, built from grouped bottleneck residual blocks, channel and spatial attention, a multi-scale depthwise hand-feature block, dual pooling, and Swish activations. Trained from scratch, it attains 96.37% accuracy (95.72% +- 0.54% over five seeds), within 1.08 percentage points of the best of nine ImageNet-pretrained efficient architectures under an identical protocol, using 8.5 to 68x fewer parameters and 1.3 to 21.7x fewer MACs. Retrained, it reaches 92.95 to 98.33% on six public BdSL benchmarks, 97.04% on a merged corpus, and 76.25% zero-shot on BdSL-38. Removing any architectural stage costs 7.61 to 89.30 points, against at most 3.17 for the training recipe. Grad-CAM with deletion-insertion and weight-randomization checks confirms that predictions follow the signing hand. A signer-independent split holding out 6 of 36 signers yields 85.18%. Quantized to 0.48 MB, it runs at 3.98 ms per image within a 15.5 MB footprint on a commodity smartphone. Together, RSBdSL38 and our from-scratch model turn benchmark accuracy into deployable accessibility at a fraction of pretrained-backbone cost; dataset, code, and models are released.

LiDAR-based place recognition is critical for long-term autonomous driving without GPS. Existing handcrafted feature methods face dual limitations. First, descriptor instability occurs due to inconsistent point cloud density from motion and environmental changes during repeated traversals. Second, representation fragility arises from reliance on single-level geometric abstractions in complex scenes. To overcome these, we propose a novel framework for 3D place recognition. We introduce an implicit 3D representation using elastic neural points. This representation is designed to reduce the influence of input-density variations and to provide more regular geometric evidence for descriptor construction. From this, we derive occupancy grids and normal vectors. These enable the construction of fused descriptors that integrate complementary perspectives: macro-level spatial layouts from a bird's-eye view and micro-scale surface geometries from 3D clusters. Extensive evaluations on diverse datasets, including KITTI, KITTI-360, MulRan, and NCLT, demonstrate that the proposed method achieves competitive and robust performance compared with representative handcrafted and learning-based baselines. The results suggest that the proposed framework provides a favorable trade-off among recognition accuracy, runtime efficiency, and map memory footprint. It also shows improved robustness under density variations and viewpoint changes in the evaluated scenarios. The code will be released soon.

ABSeeker: Training Long-Horizon Search Agents via Answer-Backtracked Credit Assignment에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

The Personalization Mirage: How LLMs Fabricate User Profiles, and Why Self-Monitoring Misleads에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

OneDayAgent: Towards a Long-Horizon Harness for Autonomous Agents에 관한 최근 업데이트입니다. 자세한 내용은 원문 링크에서 확인할 수 있습니다.

Oral health issues affect billions globally, but the cost and limited access to professional dental care hinder preventive oral healthcare. Research relies on clinical-grade radiographs or intraoral camera images, unavailable for public self-screening. This study introduces a tooth localisation and numbering model for smartphone photographs. We developed a customised Mask Region-based Convolutional Neural Network (Mask R-CNN) pipeline trained on 1,272 annotated smartphone images. To address variability in patient-generated health data, the pipeline incorporates two domain-informed mechanisms: a masked gray-world white-balancing algorithm to mitigate artificial colour casts and an anatomically constrained detection layer to enforce structural validity and suppress false positives. Evaluation comprised four stages: internal held-out testing, independent external testing, a descriptive ablation study, and fold-based training stability analysis using the same internal test set. On the internal test set, the model achieved an instance-mask AP@50 of 0.818, class-aware PQ of 0.780, and operational F1 of 0.884. Training stability showed limited between-model variation: across ten runs, instance-mask AP@50 had a standard deviation of 0.009. On the external dataset, the model achieved an instance-mask AP@50 of 0.901, class-aware PQ of 0.832, and operational F1 of 0.928 despite differences in population, sensors, and acquisition protocols. The inference pipeline is available as an open-source, containerised API. These results demonstrate that consumer-grade smartphone imagery can support automated tooth-level anatomical mapping, offering a scalable, potentially low-cost foundation for remote screening and tele-dentistry in resource-constrained environments.

The development of deep learning over the past decade has revolutionized medical imaging segmentation, allowing the extraction of precise descriptors from large volumes to characterize pathologies. Data augmentation is a technique widely regarded as a way to improve model training. It includes simple transformations like spatial operations or intensity modifications, but also more advanced synthesis techniques. Their goal is to generate new realistic samples from an existing dataset to diversify the images used during training. Among them, several propose different mixing strategies to combine real samples. However, one of their major shortcomings is to yield limited variability in terms of generated lesion shapes and locations. In this work, we introduce a novel image synthesis method, called OTLesMix, that leverages Wasserstein barycenter and optimal transport plan to generate realistic and diverse samples. We evaluated our method on three brain lesion segmentation tasks, on which it improves the Dice score compared to a model trained without synthetic data by 2.9 to 6.6 points, and outperforms state-of-the-art mix-based methods.

Current video world models struggle in multiplayer environments because they entangle world state with view-dependent visual latents, leading to redundant compute, view inconsistencies, and poor scalability. We propose MAS (Multiplayer world models with Authoritative Shared State) to resolve this limitation. Inspired by multiplayer game architectures, MAS disentangles world dynamics and view rendering. A learned Logic Engine advances a global, authoritative typed state from joint actions without any hand-written transition function, acting as the sole recurrent memory and synchronization reference. From this shared state, a learned Rendering Engine generates independent and consistent views for any requested camera on demand. This explicit disentangling allows MAS to achieve superior state accuracy and lower cross-view inconsistency compared to state-of-the-art multi-view baselines on a matched multiplayer Snake benchmark. It advances predicted worlds with 1,024 concurrent players for 10,000 recurrent steps. Our results show that explicit, authoritative state modeling provides a practical foundation for scalable and consistent multi-agent world simulation.