Search Results for "test"

Janus is a sophisticated open-source project from DeepSeek AI that aims to unify both visual understanding and image generation in a single model architecture. Rather than having separate systems for “look and describe” and “prompt and generate”, Janus uses an autoregressive transformer framework with a decoupled visual encoder—allowing it to ingest images for comprehension and to produce images from text prompts with shared internal representations. The design tackles long-standing...

MiniMax-M1 is presented as the world’s first open-weight, large-scale hybrid-attention reasoning model, designed to push the frontier of long-context, tool-using, and deeply “thinking” language models. It is built on the MiniMax-Text-01 foundation and keeps the same massive parameter budget, but reworks the attention and training setup for better reasoning and test-time compute scaling. Architecturally, it combines Mixture-of-Experts layers with lightning attention, enabling the model to support a native context length of 1 million tokens while using far fewer FLOPs than comparable reasoning models for very long generations. The team emphasizes efficient scaling of test-time compute: at 100K-token generation lengths, M1 reportedly uses only about 25 percent of the FLOPs of some competing models, making extended “think step” traces more feasible. ...

Transformer Debugger (TDB) is a research tool developed by OpenAI’s Superalignment team to investigate and interpret the behaviors of small language models. It combines automated interpretability methods with sparse autoencoders, enabling researchers to analyze how specific neurons, attention heads, and latent features contribute to a model’s outputs. TDB allows users to intervene directly in the forward pass of a model and observe how such interventions change predictions, making it...

...The project sits within a broader family of Hunyuan models that explore world generation and 3D-consistent understanding, and this mirror variant makes the reconstruction stack easier to test. It’s attractive for rapid prototyping of scenes, environment scans, or reference assets when you need repeatable 3D results from ordinary media.

MetaCLIP is a research codebase that extends the CLIP framework into a meta-learning / continual learning regime, aiming to adapt CLIP-style models to new tasks or domains efficiently. The goal is to preserve CLIP’s strong zero-shot transfer capability while enabling fast adaptation to domain shifts or novel class sets with minimal data and without catastrophic forgetting. The repository provides training logic, adaptation strategies (e.g. prompt tuning, adapter modules), and evaluation...

The CSM (Conversational Speech Model) is a speech generation model developed by Sesame AI that creates RVQ audio codes from text and audio inputs. It uses a Llama backbone and a smaller audio decoder to produce audio codes for realistic speech synthesis. The model has been fine-tuned for interactive voice demos and is hosted on platforms like Hugging Face for testing. CSM offers a flexible setup and is compatible with CUDA-enabled GPUs for efficient execution.

PRM800K is a process supervision dataset accompanying the paper Let’s Verify Step by Step, providing 800,000 step-level correctness labels on model-generated solutions to problems from the MATH dataset. The repository releases the raw labels and the labeler instructions used in two project phases, enabling researchers to study how human raters graded intermediate reasoning. Data are stored as newline-delimited JSONL files tracked with Git LFS, where each line is a full solution sample that...

The Video PreTraining (VPT) repository provides code and model artifacts for a project where agents learn to act by watching human gameplay videos—specifically, gameplay of Minecraft—using behavioral cloning. The idea is to learn general priors of control from large-scale, unlabeled video data, and then optionally fine-tune those priors for more goal-directed behavior via environment interaction. The repository contains demonstration models of different widths, fine-tuned variants (e.g. for...

...FixRes demonstrates that a mismatch between training and testing resolutions often leads to suboptimal accuracy, and fine-tuning the classifier and batch normalization layers at higher test resolutions significantly enhances performance. The repository includes pretrained models, feature embeddings, and evaluation scripts corresponding to the experiments reported in the NeurIPS 2019 paper “Fixing the train-test resolution discrepancy.”