replica
PulseAugur coverage of replica — every cluster mentioning replica across labs, papers, and developer communities, ranked by signal.
7 day(s) with sentiment data
-
AnchorComplete reconstructs complete 3D scene meshes from partial point clouds
AnchorComplete is a new framework designed to reconstruct complete 3D scene meshes from incomplete point cloud data. It addresses the common issue of partial reconstruction by integrating partial-point conditioning with…
-
New research advances 3D Gaussian Splatting for compression and real-world video
Recent research in 3D Gaussian Splatting (3DGS) focuses on improving compression, handling unsynchronized video, and enhancing feed-forward methods. GSCV++ offers lightweight compression for GS sequences by optimizing p…
-
New AOV-GS system uses foundation models for active semantic mapping
Researchers have developed AOV-GS, a novel active mapping system that leverages frozen foundation models to enhance robot exploration in human-centered environments. This system guides robots to prioritize semantically …
-
NeRF improved with self-supervised geometry regularization
Researchers have developed a new method to improve the geometric accuracy of Neural Radiance Fields (NeRF) by incorporating a self-supervised cross-ray consistency approach. This technique enhances the density field of …
-
PanoGraphGS framework unifies 3D scene graphs and panoptic segmentation
Researchers have introduced PanoGraphGS, a novel framework that unifies panoptic segmentation and hierarchical 3D scene graph construction using Gaussian splatting. This approach learns semantic and instance features on…
-
New method refines 3D Gaussian language maps for improved target localization
Researchers have developed a novel training-free method to refine 3D Gaussian language maps, enhancing their ability to locate specific user-defined targets while maintaining open-vocabulary queryability. This technique…
-
New cuVSLAM system enhances robot navigation with line features
Researchers have developed a new visual odometry system called Line-Augmented cuVSLAM, designed to improve robot navigation in environments with low texture. This system extends the existing cuVSLAM by incorporating lin…
-
New framework enhances dynamic audio-visual navigation with multi-modal learning
Researchers have introduced a new two-stage multi-modal learning framework designed to improve dynamic audio-visual navigation (D-AVN) for agents. The first stage, an Intensity-Calibrated Geometry Encoder, aligns sound …
-
AI pipeline reverse-engineers apps for learning purposes
The author used an 11-skill AI pipeline to reverse-engineer everyday applications, aiming to learn their underlying architecture rather than to replicate them. This process involved extracting knowledge from the AI to u…
-
New system ActiveLang enables robots to build language-annotated 3D maps
Researchers have developed ActiveLang, a novel system designed for autonomous 3D mapping that integrates language understanding and guided exploration. This system allows robots to build detailed 3D maps of unfamiliar e…
-
AWS MCP Server Vulnerable to Command Injection via PostgreSQL
A critical command injection vulnerability, CVE-2026-87911, has been discovered in AWS's postgres-mcp-server, allowing attackers to execute arbitrary operating system commands. The flaw, which has a CVSS score of 9.6, l…
-
New research enhances 3D Gaussian Splatting for mobile, sparse views, and dynamic scenes
Recent research explores advancements in 3D Gaussian Splatting (3DGS) to improve its efficiency, accuracy, and applicability across various scenarios. PCAsplat introduces a geometry-aware regularization framework to red…
-
DecomVoxel advances scene reconstruction with 3D-native priors
Researchers have introduced DecomVoxel, a novel framework for decompositional scene reconstruction that aims to improve the quality of reconstructed objects and backgrounds, especially in areas with significant occlusio…
-
New agentic frameworks reconstruct 3D indoor scenes for simulation
Two new research papers introduce agentic frameworks for reconstructing 3D indoor scenes, aiming to create realistic and simulation-ready digital twins. LiteReality-Agent formulates 3D reconstruction as a coding problem…
-
New CTAN framework enhances robot audio-visual navigation
Researchers have developed the Cycle-Temporal Attention Network (CTAN), a novel framework for embodied audio-visual navigation. This system aims to improve how robots integrate visual and acoustic information to locate …
-
New TDGP model enhances audio-visual navigation with adaptive replanning
Researchers have developed a new model called Transformer-based Token Fusion and Dynamic Graph Planning (TDGP) to improve audio-visual navigation for agents. This model addresses limitations in current systems by adapti…
-
VOIM system generates 3D instance maps without training
Researchers have introduced VOIM, a novel training-free system for creating 3D instance maps from RGB-D or monocular RGB data. Unlike existing online systems that commit to labels early, VOIM defers these decisions unti…
-
OccamView framework enhances 3D Gaussian reconstruction with object-aware view selection
Researchers have developed OccamView, a novel framework designed to improve active 3D Gaussian reconstruction, particularly under limited sensing budgets. This system focuses on object-conditioned view selection, using …
-
OpenBelief-Nav system enhances 3D navigation with evidence-preserving memory
Researchers have introduced OpenBelief-Nav, a novel system designed for open-vocabulary language-guided navigation in 3D environments. This system preserves observational details and reliability cues, allowing for more …
-
AI scientist 'Faraday' trained to replicate research, outperforms Claude and GPT-5.5
Researchers have developed 'Faraday,' a 27-billion-parameter AI scientist agent designed to replicate scientific research. This agent, trained using reinforcement learning with coding tools, has demonstrated superior pe…