Search papers, labs, and topics across Lattice.
Meituan Equal contribution, listed alphabeticallyCorresponding Author
11
0
11
9
Proactive agents can now be rigorously evaluated in real-world scenarios, revealing critical insights into their performance drivers.
Each evaluation run in EVA-Client not only assesses performance but also enriches the training dataset, creating a continuous feedback loop for policy improvement.
UniCoder achieves state-of-the-art visual-to-code generation by transforming blind exploration into guided policy improvement, setting a new standard for the field.
MLLMs struggle with video temporal-logical reasoning, showing a substantial performance gap compared to human capabilities, especially as complexity increases.
InterleaveThinker transforms standard image generators into powerful tools for interleaved generation, achieving state-of-the-art performance while enhancing reasoning capabilities.
LLMs can now build playable games from scratch, thanks to a new framework that teaches them to scaffold stable architectures and systematically debug integration errors, not just patch syntax.
By tightly coupling reasoning, searching, and generation, Unify-Agent demonstrates that agent-based modeling can substantially improve world knowledge grounding in image synthesis, rivaling closed-source models.
Image generation takes a leap towards real-world knowledge by training an agent that actively searches for and integrates external information, substantially boosting performance on knowledge-intensive tasks.
LongCat-Next shatters the language-centric paradigm by unifying text, vision, and audio into a single autoregressive model with minimal modality-specific design, finally reconciling understanding and generation in discrete vision modeling.
Achieve photorealistic and structurally consistent weather editing for autonomous driving videos without the massive datasets typically required by generative models.
Representation-Pivoted Autoencoders enable diffusion models to generate and edit images with higher fidelity by learning a compressed latent space that preserves the semantics of pre-trained visual representations.
