📌 TOPINDIATOURS Update ai: Are quantum particles polygamous? Electron crowding tri

Are quantum particles polygamous? New experiments suggest some of them abandon long-standing partnerships when conditions get crowded.

Quantum particles do not behave like isolated dots.

They interact, form bonds, and follow strict social rules. One of the most fundamental divides separates fermions and bosons.

Fermions refuse to share quantum states. Bosons happily pile together.

Those opposing traits underpin everything from solid matter to superconductors.

But new work shows that quantum relationships can break down in unexpected ways.

Researchers found that under extreme conditions, particles once thought to be strictly “monogamous” can suddenly change partners.

The result flips long-held assumptions about how particles move through materials.

When quantum norms fail

Electrons sometimes bind tightly to atoms, locking a material into an insulating state.

In other cases, they roam freely and carry electric current.

Under special conditions, electrons even pair with each other into Cooper pairs, enabling superconductivity.

Another important pairing involves electrons and holes.

A hole forms when an atom in a material loses an electron, leaving behind a mobile positive charge.

When an electron and hole bind, they create an exciton. Physicists often describe excitons as monogamous because breaking them apart requires energy.

Excitons behave like bosons. Individual electrons remain fermions.

That contrast makes them ideal for studying how fermions and bosons interact.

JQI Fellow Mohammad Hafezi and his colleagues wanted to see how changing the balance between these particles affects motion inside a material.

They predicted that packing a material with fermionic electrons would block excitons and slow them down.

The experiment produced the opposite result.

“We thought the experiment was done wrong,” says Daniel Suárez-Forero, a former JQI postdoctoral researcher who is now an assistant professor at the University of Maryland, Baltimore County. “That was the first reaction.”

The team built a carefully aligned layered material. Its structure forced electrons and excitons into a tidy grid of allowed positions. Electrons refused to share those sites. Excitons could hop between them.

At low electron densities, excitons behaved normally.

As more electrons entered the system, exciton motion slowed. Their paths became indirect as they navigated around occupied sites.

Then the system crossed a threshold.

When nearly every site filled with an electron, exciton mobility jumped sharply. Instead of freezing, excitons suddenly traveled farther than before.

“No one wanted to believe it,” says Pranshoo Upadhyay, a JQI graduate student and lead author of the paper.

‘It’s like, can you repeat it? And for about a month, we performed measurements on different locations of the sample with different excitation powers and replicated it in several other samples.”

The team repeated the experiment across samples, setups, and even continents. The effect persisted.

“We repeated the experiment in a different sample, in a different setup, and even in a different continent, and the result was exactly the same,” Suárez-Forero says.

Beyond exciton monogamy

Theory eventually caught up with experiment.

The researchers realized that excitons did not sit in the same way as free electrons and holes.

“At least this was what we thought,” said Tsung-Sheng Huang, a former JQI graduate student of the group who is now a postdoctoral researcher at the Institute of Photonic Sciences in Spain.

“Any external fermion should not see the constituents of the exciton separately; but in reality, the story is a little bit different.”

At very high electron densities, holes inside excitons began treating all nearby electrons as equivalent. The exclusive bond broke down.

Holes effectively switched partners repeatedly, a process the team calls non-monogamous hole diffusion.

That rapid partner-switching allowed excitons to move straight through the crowded system.

Instead of weaving around obstacles, they traveled efficiently before recombining and emitting light.

The researchers triggered the effect simply by adjusting the voltage.

That control makes the phenomenon attractive for future electronic and optical devices, including exciton-based solar technologies.

The study is published in the journal Science.

🔗 Sumber: interestingengineering.com

📌 TOPINDIATOURS Hot ai: ByteDance Introduces Astra: A Dual-Model Architecture for

The increasing integration of robots across various sectors, from industrial manufacturing to daily life, highlights a growing need for advanced navigation systems. However, contemporary robot navigation systems face significant challenges in diverse and complex indoor environments, exposing the limitations of traditional approaches. Addressing the fundamental questions of “Where am I?”, “Where am I going?”, and “How do I get there?”, ByteDance has developed Astra, an innovative dual-model architecture designed to overcome these traditional navigation bottlenecks and enable general-purpose mobile robots.

Traditional navigation systems typically consist of multiple, smaller, and often rule-based modules to handle the core challenges of target localization, self-localization, and path planning. Target localization involves understanding natural language or image cues to pinpoint a destination on a map. Self-localization requires a robot to determine its precise position within a map, especially challenging in repetitive environments like warehouses where traditional methods often rely on artificial landmarks (e.g., QR codes). Path planning further divides into global planning for rough route generation and local planning for real-time obstacle avoidance and reaching intermediate waypoints.

While foundation models have shown promise in integrating smaller models to tackle broader tasks, the optimal number of models and their effective integration for comprehensive navigation remained an open question.

ByteDance’s Astra, detailed in their paper “Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning” (website: https://astra-mobility.github.io/), addresses these limitations. Following the System 1/System 2 paradigm, Astra features two primary sub-models: Astra-Global and Astra-Local. Astra-Global handles low-frequency tasks like target and self-localization, while Astra-Local manages high-frequency tasks such as local path planning and odometry estimation. This architecture promises to revolutionize how robots navigate complex indoor spaces.

Astra-Global: The Intelligent Brain for Global Localization

Astra-Global serves as the intelligent core of the Astra architecture, responsible for critical low-frequency tasks: self-localization and target localization. It functions as a Multimodal Large Language Model (MLLM), adept at processing both visual and linguistic inputs to achieve precise global positioning within a map. Its strength lies in utilizing a hybrid topological-semantic graph as contextual input, allowing the model to accurately locate positions based on query images or text prompts.

The construction of this robust localization system begins with offline mapping. The research team developed an offline method to build a hybrid topological-semantic graph G=(V,E,L):

• V (Nodes): Keyframes, obtained by temporal downsampling of input video and SfM-estimated 6-Degrees-of-Freedom (DoF) camera poses, act as nodes encoding camera poses and landmark references.
• E (Edges): Undirected edges establish connectivity based on relative node poses, crucial for global path planning.
• L (Landmarks): Semantic landmark information is extracted by Astra-Global from visual data at each node, enriching the map’s semantic understanding. These landmarks store semantic attributes and are connected to multiple nodes via co-visibility relationships.

In practical localization, Astra-Global’s self-localization and target localization capabilities leverage a coarse-to-fine two-stage process for visual-language localization. The coarse stage analyzes input images and localization prompts, detects landmarks, establishes correspondence with a pre-built landmark map, and filters candidates based on visual consistency. The fine stage then uses the query image and coarse output to sample reference map nodes from the offline map, comparing their visual and positional information to directly output the predicted pose.

For language-based target localization, the model interprets natural language instructions, identifies relevant landmarks using their functional descriptions within the map, and then leverages landmark-to-node association mechanisms to locate relevant nodes, retrieving target images and 6-DoF poses.

To empower Astra-Global with robust localization abilities, the team employed a meticulous training methodology. Using Qwen2.5-VL as the backbone, they combined Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). SFT involved diverse datasets for various tasks, including coarse and fine localization, co-visibility detection, and motion trend estimation. In the GRPO phase, a rule-based reward function (including format, landmark extraction, map matching, and extra landmark rewards) was used to train for visual-language localization. Experiments showed GRPO significantly improved Astra-Global’s zero-shot generalization, achieving 99.9% localization accuracy in unseen home environments, surpassing SFT-only methods.

Astra-Local: The Intelligent Assistant for Local Planning

Astra-Local acts as the intelligent assistant for Astra’s high-frequency tasks, a multi-task network capable of efficiently generating local paths and accurately estimating odometry from sensor data. Its architecture comprises three core components: a 4D spatio-temporal encoder, a planning head, and an odometry head.

The 4D spatio-temporal encoder replaces traditional mobile stack perception and prediction modules. It begins with a 3D spatial encoder that processes N omnidirectional images through a Vision Transformer (ViT) and Lift-Splat-Shoot to convert 2D image features into 3D voxel features. This 3D encoder is trained using self-supervised learning via 3D volumetric differentiable neural rendering. The 4D spatio-temporal encoder then builds upon the 3D encoder, taking past voxel features and future timestamps as input to predict future voxel features through ResNet and DiT modules, providing current and future environmental representations for planning and odometry.

The planning head, based on pre-trained 4D features, robot speed, and task information, generates executable trajectories using Transformer-based flow matching. To prevent collisions, the planning head incorporates a masked ESDF loss (Euclidean Signed Distance Field). This loss calculates the ESDF of a 3D occupancy map and applies a 2D ground truth trajectory mask, significantly reducing collision rates. Experiments demonstrate its superior performance in collision rate and overall score on out-of-distribution (OOD) datasets compared to other methods.

The odometry head predicts the robot’s relative pose using current and past 4D features and additional sensor data (e.g., IMU, wheel data). It trains a Transformer model to fuse information from different sensors. Each sensor modality is processed by a specific tokenizer, combined with modality embeddings and temporal positional embeddi…

Konten dipersingkat otomatis.

🔗 Sumber: syncedreview.com