Engineering
DramaChain Bench: An End-to-End Benchmark for Short-Drama Generation
DramaChain Bench is a benchmark for the complete short-drama production pipeline, spanning script, storyboard, keyframes, shots, and final episodes. It…
Engineering
Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement
This paper finds that on-policy distillation’s teacher supervision contains substantial noise that doesn’t meaningfully affect student performance, and proposes On-Policy…
Engineering
Lucida: Parse, Generate, and Place for Composable Real-to-Sim Scene Modeling
Lucida converts real indoor scenes into editable 3D simulation environments through a three-stage pipeline: parsing video into scene graphs with…
Engineering
DreamX-Creator: Democratizing Native Audio-Video Generation at 2K Resolution
DreamX-Creator 1.0 is a compact 7-billion-parameter system for unified audio-video generation that jointly processes both modalities using gated cross-modal attention.…
Engineering
GenFirst: Generation Before Reconstruction for Stable End-to-End Latent Generative Modeling
GenFirst addresses latent collapse in end-to-end training of latent generative models by reversing the usual two-stage pipeline: a generation-before-reconstruction strategy…
Engineering
Normalized Low-Rank Adaptation
Normalized Low-Rank Adaptation (NoRA) normalizes the down-projection matrices during LoRA training to improve optimization stability, and can also be applied…
Engineering
PaperGym: Rubric-Centered Evolution for Research-Plan Generation
PaperGym converts research papers into AI training environments by extracting evaluation criteria from a paper’s methodology and experiments rather than…
Engineering
CogEvol: Towards Efficient and Reliable Learning Environment Generation
CogEvol is a family of models that automatically generate learning environments by converting course briefs into structured JSON slides or…
Engineering
SHAPE of Chain-of-Thought in Math Reasoning
SHAPE is an analytical framework examining chain-of-thought reasoning in LLMs through two mathematics-education lenses: semantic spaces (different mathematical interpretations) and…
Engineering
LightNav-0: Eliciting VLM Spatial Intelligence for Generalist Embodied Navigation
LightNav-0 is a compact embodied-navigation model that leverages the spatial reasoning of pretrained vision-language models through a unified interface combining…
Engineering
Super Library Agent: Joint Generation and Maintenance of Multiple Applications Beyond the Single Codebase
Super Library Agent lets LLM coding agents sequentially generate multiple related applications while maintaining a shared library of reusable components,…
Engineering
Evaluating the Hidden Costs of Personalization in Large Language Models
This paper introduces PRISK, a dynamic evaluation framework with automated data generation and tailored metrics that systematically assesses unintended consequences…