CCTest · Blog
Distilled RL brings teacher guidance into reinforcement learning for LLM post-training
Reinforcement Learning
cctest.ai

Distilled RL brings teacher guidance into reinforcement learning for LLM post-training

Distilled Reinforcement Learning proposes a middle path between outcome-only RL and unconditional online policy distillation. It uses a teacher model to redistribute token-level policy-gradient signals while keeping optimization anchored to rewards.

Read more
CCTest · Blog
DiffGI Reframes Thin-Shell 3D Generation as Differentiable Geometry Images
Diffusion Models
cctest.ai
Diffusion Models

DiffGI Reframes Thin-Shell 3D Generation as Differentiable Geometry Images

DiffGI targets a long-standing weakness of many 3D generative models: representing thin-shell and non-manifold surfaces such as garments. By combining continuous 2D TSDF maps with differentiable Marching Squares, it links compact 2D latent representations directly to 3D surface optimization.

Read more
CCTest · Blog
Xiaomi-Robotics-1 Brings Scaling Laws to Robot VLA Models with 100K+ Hours of Real Trajectories
Robotics & Physical AI
cctest.ai

Xiaomi-Robotics-1 Brings Scaling Laws to Robot VLA Models with 100K+ Hours of Real Trajectories

Xiaomi-Robotics-1 explores whether robotics can benefit from the same scaling dynamics that transformed language and vision models. By pre-training on more than 100,000 hours of real manipulation trajectories, the model shows consistent gains as data and model size increase.

Read more
CCTest · Blog
S1-Omni Unifies Scientific Multimodal Reasoning Across Molecules, Proteins, Materials, and Images
AI for Science
cctest.ai
AI for Science

S1-Omni Unifies Scientific Multimodal Reasoning Across Molecules, Proteins, Materials, and Images

S1-Omni aims to reduce fragmentation in AI for Science by bringing diverse scientific data types into one reasoning model. It supports more than 200 scientific tasks and releases model weights, inference code, and a 10K-sample corpus subset.

Read more
CCTest · Blog
Robot-Centric Pointmaps Help VLA Models See From the Robot’s Frame
Robotics & Physical AI
cctest.ai

Robot-Centric Pointmaps Help VLA Models See From the Robot’s Frame

A KAIST AI paper introduces robot-centric pointmaps to reduce the mismatch between camera observations and robot-frame actions in VLA models. The method encodes 3D scene coordinates in the robot frame while keeping the image-like grid structure used by existing 2D vision backbones.

Read more
CCTest · Blog
RecGPT-V3: Moving Recommenders from Replaying History to Stateful Intent Understanding
Memory & Context
cctest.ai
Memory & Context

RecGPT-V3: Moving Recommenders from Replaying History to Stateful Intent Understanding

RecGPT-V3 presents an industrial LLM-based recommender that maintains evolving user memory, grounds intent with Semantic IDs, and compresses reasoning into latent tokens. In Taobao’s “Guess What You Like” feed, the report cites higher engagement and a 52.4% reduction in serving resource consumption.

Read more