Discover the Best AI Resources

datapointai/text-2-image-human-preferences-2m

Contains ~2 million human pairwise preference judgments comparing images generated from text prompts; each example pairs two images with a preferred/tie label and is formatted for preference learning, reward-model training, and evaluation.

FACET: Preserving Source Intent and Executable State in Terminal Task Synthesis

Proposes FACET, a framework that synthesizes verifiable terminal tasks by reconstructing scenario intent and grounding instruction, solution, and verifier in a shared executable container state. Key features include environment-first generation, execution-based validation, and targeted repair to preserve source intent and cross-artifact consistency.

EnvHarness: Awakening Static Worlds for Agent Learning

Wraps static, hand-built environments with a programmable plug-in harness that reshapes environment behavior without changing underlying logic. EnvRigger automates diagnosis and synthesis of harness components from agent failure trajectories, validating edits via fresh rollouts to improve agent success and efficiency.

SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?

Evaluates whether coding agents can modify real scientific software while preserving domain-specific scientific contracts. Contains 119 repository-level tasks across 98 GitHub projects and 20 scientific domains, measures reproducible edits in pinned Docker images, and analyzes recurring failure modes.

Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts

Estimates optimal learning rates for large-scale Mixture-of-Experts pretraining using a two-step, compute-efficient transfer: μP-based width transfer from small proxy models, then log-log linear extrapolation across token budgets to trillion-token horizons.

Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs

Treats human annotations as oracle rollouts and separates them from on-policy baselines to improve reinforcement learning for video multimodal LLMs. Key features include a decoupled advantage estimator, sign-balanced pruning, and scalable gains across model sizes and data budgets.

VGI-Bench: Probing Visual Intelligence in Video Generation Models

Evaluates visual reasoning in video generation models using 27 photorealistic tasks (810 instances), a two-level taxonomy of domains and skill tags, and task designs that enforce valid intermediate trajectories and calibrated difficulty.

Qwen3.8-27B-Escha-W2

Provides 2-bit quantized weights of Qwen3.8-27B (~10.15 GB) for local deployment, enabling the full 27B parameter model to run on a single 24 GB GPU with long-context support. Delivered as safetensors plus a companion SGLang runtime; measured to match FP8 reference on common benchmarks with small or no quality loss.

4DAnyone: Create Anyone in 4D from a Casual Monocular Video

Turns an uncalibrated monocular actor video into multiview-consistent novel-view videos and lifts them into 4D Gaussian Splatting assets. Introduces Reference Context Packing to keep reference conditioning fixed-size and Target Context Routing to exchange context across target groups, improving large-view reconstruction consistency.

WithEveryone: Unified Planning and Identity Grounding for Group Image Generation

Generates group images that bind up to ten reference identities to distinct people and locations by predicting an explicit identity–layout plan and supervising faces with Layout-Grounded ID Loss. Improves identity fidelity while cutting copy-paste duplication; suited for multi-person image synthesis but requires identity-annotated face regions and paired training data.

CAD 1000 Hours

Provides 1,021.64 hours across 597 CAD/BIM workflows with synchronized screen recordings and interaction logs; each workflow includes video, timestamped input events, task specs, source files, final outputs, and evaluation rubrics for training or evaluating desktop CAD agents.

InfinityEdit: Infinite Video Editing with a Lightweight Edit-Ignition Adapter

Continues a live or ongoing video stream while applying user-specified edits on the fly using a lightweight edit-ignition adapter. The adapter injects edits only in chunks where requests arrive and uses history cross-attention and temporal causal self-attention to preserve continuity and stability for unbounded streaming edits.